Merge pull request #1262 from xinnan-tech/web-headerbar-modify

update:修复init_memorybug

README.md

@@ -10,7 +10,7 @@
 </p>
 
 <p align="center">
-<a href="./README.md">English</a>
+<a href="./README_en.md">English</a>
 · <a href="./docs/FAQ.md">常见问题</a>
 · <a href="https://github.com/xinnan-tech/xiaozhi-esp32-server/issues">反馈问题</a>
 · <a href="./README.md#%E9%83%A8%E7%BD%B2%E6%96%87%E6%A1%A3">部署文档</a>

README_en.md

@@ -0,0 +1,273 @@
+[![Banners](docs/images/banner1.png)](https://github.com/xinnan-tech/xiaozhi-esp32-server)
+
+<h1 align="center">Xiaozhi Backend Service xiaozhi-esp32-server</h1>
+
+<p align="center">
+This project provides backend services for the open-source smart hardware project
+<a href="https://github.com/78/xiaozhi-esp32">xiaozhi-esp32</a><br/>
+Implemented using Python, Java, and Vue according to the <a href="https://ccnphfhqs21z.feishu.cn/wiki/M0XiwldO9iJwHikpXD5cEx71nKh">Xiaozhi Communication Protocol</a><br/>
+Helping you quickly set up your Xiaozhi server
+</p>
+
+<p align="center">
+<a href="./README.md">中文</a>
+· <a href="./docs/FAQ.md">FAQ</a>
+· <a href="https://github.com/xinnan-tech/xiaozhi-esp32-server/issues">Report Issues</a>
+· <a href="./README_ed.md#deployment-documentation">Deployment Guide</a>
+· <a href="https://github.com/xinnan-tech/xiaozhi-esp32-server/releases">Release Notes</a>
+</p>
+<p align="center">
+  <a href="https://github.com/xinnan-tech/xiaozhi-esp32-server/releases">
+    <img alt="GitHub Contributors" src="https://img.shields.io/github/v/release/xinnan-tech/xiaozhi-esp32-server?logo=docker" />
+  </a>
+  <a href="https://github.com/xinnan-tech/xiaozhi-esp32-server/graphs/contributors">
+    <img alt="GitHub Contributors" src="https://img.shields.io/github/contributors/xinnan-tech/xiaozhi-esp32-server?logo=github" />
+  </a>
+  <a href="https://github.com/xinnan-tech/xiaozhi-esp32-server/issues">
+    <img alt="Issues" src="https://img.shields.io/github/issues/xinnan-tech/xiaozhi-esp32-server?color=0088ff" />
+  </a>
+  <a href="https://github.com/xinnan-tech/xiaozhi-esp32-server/pulls">
+    <img alt="GitHub pull requests" src="https://img.shields.io/github/issues-pr/xinnan-tech/xiaozhi-esp32-server?color=0088ff" />
+  </a>
+  <a href="https://github.com/xinnan-tech/xiaozhi-esp32-server/blob/main/LICENSE">
+    <img alt="GitHub pull requests" src="https://img.shields.io/badge/license-MIT-white?labelColor=black" />
+  </a>
+  <a href="https://github.com/xinnan-tech/xiaozhi-esp32-server">
+    <img alt="stars" src="https://img.shields.io/github/stars/xinnan-tech/xiaozhi-esp32-server?color=ffcb47&labelColor=black" />
+  </a>
+</p>
+
+---
+
+## Target Users 👥
+
+This project requires ESP32 hardware devices. If you have purchased ESP32-related hardware, successfully connected to Brother Xia's backend service, and want to set up your own `xiaozhi-esp32` backend service, then this project is perfect for you.
+
+Want to see it in action? Check out these videos 🎥
+
+<table>
+  <tr>
+    <td>
+        <a href="https://www.bilibili.com/video/BV1FMFyejExX" target="_blank">
+         <picture>
+           <img alt="Xiaozhi esp32 connecting to custom backend model" src="docs/images/demo1.png" />
+         </picture>
+        </a>
+    </td>
+    <td>
+        <a href="https://www.bilibili.com/video/BV1CDKWemEU6" target="_blank">
+         <picture>
+           <img alt="Custom voice" src="docs/images/demo2.png" />
+         </picture>
+        </a>
+    </td>
+    <td>
+        <a href="https://www.bilibili.com/video/BV12yA2egEaC" target="_blank">
+         <picture>
+           <img alt="Cantonese communication" src="docs/images/demo3.png" />
+         </picture>
+        </a>
+    </td>
+    <td>
+        <a href="https://www.bilibili.com/video/BV1pNXWYGEx1" target="_blank">
+         <picture>
+           <img alt="Home appliance control" src="docs/images/demo5.png" />
+         </picture>
+        </a>
+    </td>
+    <td>
+        <a href="https://www.bilibili.com/video/BV1kgA2eYEQ9" target="_blank">
+         <picture>
+           <img alt="Lowest cost configuration" src="docs/images/demo4.png" />
+         </picture>
+        </a>
+    </td>
+  </tr>
+  <tr>
+    <td>
+        <a href="https://www.bilibili.com/video/BV1Vy96YCE3R" target="_blank">
+         <picture>
+           <img alt="Custom voice" src="docs/images/demo6.png" />
+         </picture>
+        </a>
+    </td>
+    <td>
+        <a href="https://www.bilibili.com/video/BV1VC96Y5EMH" target="_blank">
+         <picture>
+           <img alt="Music playback" src="docs/images/demo7.png" />
+         </picture>
+        </a>
+    </td>
+    <td>
+        <a href="https://www.bilibili.com/video/BV1Z8XuYZEAS" target="_blank">
+         <picture>
+           <img alt="Weather plugin" src="docs/images/demo8.png" />
+         </picture>
+        </a>
+    </td>
+    <td>
+        <a href="https://www.bilibili.com/video/BV178XuYfEpi" target="_blank">
+         <picture>
+           <img alt="IOT device control" src="docs/images/demo9.png" />
+         </picture>
+        </a>
+    </td>
+    <td>
+        <a href="https://www.bilibili.com/video/BV17LXWYvENb" target="_blank">
+         <picture>
+           <img alt="News broadcast" src="docs/images/demo0.png" />
+         </picture>
+        </a>
+    </td>
+  </tr>
+</table>
+
+---
+
+## Warning ⚠️
+
+1. This project is open-source software. This software has no commercial relationship with any third-party API service providers (including but not limited to speech recognition, large models, speech synthesis, and other platforms) and does not provide any form of guarantee for their service quality or financial security.
+It is recommended that users prioritize service providers with relevant business licenses and carefully read their service agreements and privacy policies. This software does not host any account keys, does not participate in fund transfers, and does not bear the risk of recharge fund losses.
+
+2. This project's functionality is not complete and has not passed network security testing. Please do not use it in production environments. If you deploy this project for learning in a public network environment, please ensure necessary protection measures are in place.
+
+---
+
+## Deployment Documentation
+
+![Banners](docs/images/banner2.png)
+
+This project offers two deployment methods. Please choose based on your specific needs:
+
+#### 🚀 Deployment Method Selection
+
+| Deployment Method | Features | Use Case | Docker Deployment Guide | Source Code Deployment Guide |
+|---------|------|---------|---------|---------|
+| **Simplified Installation** | Smart dialogue, IOT functionality, data stored in configuration files | Low-configuration environment, no database required | [Docker Server Only](./docs/Deployment.md#%E6%96%B9%E5%BC%8F%E4%B8%80docker%E5%8F%AA%E8%BF%90%E8%A1%8Cserver) | [Local Source Code Server Only](./docs/Deployment.md#%E6%96%B9%E5%BC%8F%E4%BA%8C%E6%9C%AC%E5%9C%B0%E6%BA%90%E7%A0%81%E5%8F%AA%E8%BF%90%E8%A1%8Cserver)|
+| **Full Module Installation** | Smart dialogue, IOT, OTA, Control Panel, data stored in database | Complete functionality experience |[Docker Full Module](./docs/Deployment_all.md#%E6%96%B9%E5%BC%8F%E4%B8%80docker%E8%BF%90%E8%A1%8C%E5%85%A8%E6%A8%A1%E5%9D%97) | [Local Source Code Full Module](./docs/Deployment_all.md#%E6%96%B9%E5%BC%8F%E4%BA%8C%E6%9C%AC%E5%9C%B0%E6%BA%90%E7%A0%81%E8%BF%90%E8%A1%8C%E5%85%A8%E6%A8%A1%E5%9D%97) |
+
+> 💡 Note: Below is the test platform deployed with the latest code. You can flash and test if needed. Concurrent users: 6, data cleared daily
+
+```
+Control Panel: https://2662r3426b.vicp.fun
+
+Service Test Tool: https://2662r3426b.vicp.fun/test/
+OTA Interface: https://2662r3426b.vicp.fun/xiaozhi/ota/
+Websocket Interface: wss://2662r3426b.vicp.fun/xiaozhi/v1/
+```
+
+---
+## Feature List ✨
+
+### Implemented ✅
+
+| Feature Module | Description |
+|---------|------|
+| Communication Protocol | Based on `xiaozhi-esp32` protocol, implements data interaction through WebSocket |
+| Dialogue Interaction | Supports wake-up dialogue, manual dialogue, and real-time interruption. Auto-sleep after long periods of inactivity |
+| Intent Recognition | Supports LLM intent recognition, function call, reducing hard-coded intent judgment |
+| Multi-language Recognition | Supports Mandarin, Cantonese, English, Japanese, Korean (default using FunASR) |
+| LLM Module | Supports flexible LLM module switching, default using ChatGLMLLM, also supports Ali Bailing, DeepSeek, Ollama, etc. |
+| TTS Module | Supports EdgeTTS (default), Volcano Engine Doubao TTS, and other TTS interfaces for speech synthesis |
+| Memory Function | Supports ultra-long memory, local summary memory, and no memory modes for different scenarios |
+| IOT Function | Supports managing registered device IOT functionality, intelligent IoT control based on dialogue context |
+| Control Panel | Provides web management interface, supports agent management, user management, system configuration, etc. |
+
+### In Development 🚧
+
+To learn about specific development progress, [click here](https://github.com/users/xinnan-tech/projects/3)
+
+If you're a software developer, here's an [Open Letter to Developers](docs/contributor_open_letter.md). Welcome to join!
+
+---
+
+## Product Ecosystem 👬
+Xiaozhi is an ecosystem. When using this product, you might want to check out other excellent projects in this ecosystem:
+
+| Project Name | Project Link | Description |
+|:---------------------|:--------|:--------|
+| Xiaozhi Android Client | [xiaozhi-android-client](https://github.com/TOM88812/xiaozhi-android-client) | A Flutter-based Android and iOS voice dialogue application supporting real-time voice interaction and text dialogue |
+| Xiaozhi PC Client | [py-xiaozhi](https://github.com/Huang-junsen/py-xiaozhi) | A Python-based AI client that allows you to experience Xiaozhi AI functionality through code without physical hardware |
+| Xiaozhi Java Server | [xiaozhi-esp32-server-java](https://github.com/joey-zhou/xiaozhi-esp32-server-java) | A Java-based open-source project providing complete backend service solutions |
+
+---
+
+## Supported Platforms/Components 📋
+
+### LLM Language Models
+
+| Usage Method | Supported Platforms | Free Platforms |
+|:---:|:---:|:---:|
+| openai API | Ali Bailing, Volcano Engine Doubao, DeepSeek, ChatGLM, Gemini | ChatGLM, Gemini |
+| ollama API | Ollama | - |
+| dify API | Dify | - |
+| fastgpt API | Fastgpt | - |
+| coze API | Coze | - |
+
+Actually, any LLM supporting openai API calls can be integrated.
+
+---
+
+### TTS Speech Synthesis
+
+| Usage Method | Supported Platforms | Free Platforms |
+|:---:|:---:|:---:|
+| API Calls | EdgeTTS, Volcano Engine Doubao TTS, Tencent Cloud, Aliyun TTS, CosyVoiceSiliconflow, TTS302AI, CozeCnTTS, GizwitsTTS, ACGNTTS, OpenAITTS | EdgeTTS, CosyVoiceSiliconflow(partial) |
+| Local Service | FishSpeech, GPT_SOVITS_V2, GPT_SOVITS_V3, MinimaxTTS | FishSpeech, GPT_SOVITS_V2, GPT_SOVITS_V3, MinimaxTTS |
+
+---
+
+### VAD Voice Activity Detection
+
+| Type | Platform Name | Usage Method | Pricing | Notes |
+|:---:|:---------:|:----:|:----:|:--:|
+| VAD | SileroVAD | Local Use | Free | |
+
+---
+
+### ASR Speech Recognition
+
+| Usage Method | Supported Platforms | Free Platforms |
+|:---:|:---:|:---:|
+| Local Use | FunASR, SherpaASR | FunASR, SherpaASR |
+| API Calls | DoubaoASR, FunASRServer, TencentASR, AliyunASR | FunASRServer |
+
+---
+
+### Memory Storage
+
+| Type | Platform Name | Usage Method | Pricing | Notes |
+|:------:|:---------------:|:----:|:---------:|:--:|
+| Memory | mem0ai | API Calls | 1000 calls/month quota | |
+| Memory | mem_local_short | Local Summary | Free | |
+
+---
+
+### Intent Recognition
+
+| Type | Platform Name | Usage Method | Pricing | Notes |
+|:------:|:-------------:|:----:|:-------:|:---------------------:|
+| Intent | intent_llm | API Calls | Based on LLM pricing | Uses large model for intent recognition, highly versatile |
+| Intent | function_call | API Calls | Based on LLM pricing | Uses large model function calls for intent, fast and effective |
+
+---
+
+## Acknowledgments 🙏
+
+| Logo | Project/Company | Description |
+|:---:|:---:|:---|
+| <img src="./docs/images/logo_bailing.png" width="160"> | [Bailing Voice Dialogue Robot](https://github.com/wwbin2017/bailing) | This project was inspired by [Bailing Voice Dialogue Robot](https://github.com/wwbin2017/bailing) and implemented based on it |
+| <img src="./docs/images/logo_tenclass.png" width="160"> | [Tenclass](https://www.tenclass.com/) | Thanks to [Tenclass](https://www.tenclass.com/) for developing standard communication protocols, multi-device compatibility solutions, and high-concurrency scenario practices for the Xiaozhi ecosystem; providing comprehensive technical documentation support for this project |
+| <img src="./docs/images/logo_xuanfeng.png" width="160"> | [Xuanfeng Technology](https://github.com/Eric0308) | Thanks to [Xuanfeng Technology](https://github.com/Eric0308) for contributing function call framework, MCP communication protocol, and plugin call mechanism implementation code, significantly improving front-end device (IoT) interaction efficiency and functional extensibility through standardized instruction scheduling system and dynamic expansion capabilities |
+| <img src="./docs/images/logo_huiyuan.png" width="160"> | [Huiyuan Design](http://ui.kwd988.net/) | Thanks to [Huiyuan Design](http://ui.kwd988.net/) for providing professional visual solutions for this project, empowering product user experience with their design experience serving over a thousand enterprises |
+| <img src="./docs/images/logo_qinren.png" width="160"> | [Xi'an Qinren Information Technology](https://www.029app.com/) | Thanks to [Xi'an Qinren Information Technology](https://www.029app.com/) for deepening this project's visual system, ensuring consistency and extensibility of overall design style in multi-scenario applications |
+
+
+<a href="https://star-history.com/#xinnan-tech/xiaozhi-esp32-server&Date">
+
+ <picture>
+   <source media="(prefers-color-scheme: dark)" srcset="https://api.star-history.com/svg?repos=xinnan-tech/xiaozhi-esp32-server&type=Date&theme=dark" />
+   <source media="(prefers-color-scheme: light)" srcset="https://api.star-history.com/svg?repos=xinnan-tech/xiaozhi-esp32-server&type=Date" />
+   <img alt="Star History Chart" src="https://api.star-history.com/svg?repos=xinnan-tech/xiaozhi-esp32-server&type=Date" />
+ </picture>
+</a> 

docs/fish-speech-integration.md

@@ -0,0 +1,72 @@
+登录AutoDL，租赁镜像
+选择镜像:
+```
+PyTorch / 2.1.0 / 3.10(ubuntu22.04) / cuda 12.1
+```
+
+机器开机后，设置学术加速
+```
+source /etc/network_turbo
+```
+
+进入工作目录
+```
+cd autodl-tmp/
+```
+
+拉取项目
+```
+git clone https://gitclone.com/github.com/fishaudio/fish-speech.git ; cd fish-speech
+```
+
+安装依赖
+```
+pip install -e.
+```
+
+如果报错，安装portaudio
+```
+apt-get install portaudio19-dev -y
+```
+
+安装后执行
+```
+pip install torch==2.3.1 torchvision==0.18.1 torchaudio==2.3.1 --index-url https://download.pytorch.org/whl/cu121
+```
+
+下载模型
+```
+cd tools
+python download_models.py 
+```
+
+下载完模型后运行接口
+```
+python -m tools.api_server --listen 0.0.0.0:6006 
+```
+
+然后用浏览器去到aotodl实例页面
+```
+https://autodl.com/console/instance/list
+```
+
+如下图点击你刚才机器的`自定义服务`按钮，开启端口转发服务
+![自定义服务](images/fishspeech/autodl-01.png)
+
+端口转发服务设置完成后，你本地电脑打开网址`http://localhost:6006/`，就可以访问fish-speech的接口了
+![服务预览](images/fishspeech/autodl-02.png)
+
+
+如果你是单模块部署，核心配置如下
+```
+selected_module:
+  TTS: FishSpeech
+TTS:
+  FishSpeech:
+    reference_audio: ["config/assets/wakeup_words.wav",]
+    reference_text: ["哈啰啊，我是小智啦，声音好听的台湾女孩一枚，超开心认识你耶，最近在忙啥，别忘了给我来点有趣的料哦，我超爱听八卦的啦",]
+    api_key: "123"
+    api_url: "http://127.0.0.1:6006/v1/tts"
+```
+
+然后重启服务

docs/homeassistant-integration.md

@@ -1,4 +1,8 @@
-# ESP32设备与HomeAssistant集成指南
+# 小智ESP32-开源服务端与HomeAssistant集成指南
+
+[TOC]
+
+-----
 
 ## 简介
 
@@ -7,26 +11,58 @@
 ## 前提条件
 
 - 已安装并配置好`HomeAssistant`
-- 已经在HomeAssistant中集成`Xiaomi Home`，并将米家的设备导入进`HomeAssistant`
-- `xiaozhi-esp32-server智控台`能正常使用
 - 本次我选择的模型是：免费的ChatGLM，它支持functioncall函数调用
 
-## 网络环境
+## 开始前的操作（必要）
+
+### 1. 获取HA的网络网络地址信息
+
+请访问你Home Assistant的网络地址，例如，我的HA的地址是192.168.4.7，端口则是默认的8123，则在浏览器打开
 
-1、我的`HomeAssistant`部署在下面，版本是`2025.3.4`
 ```
 http://192.168.4.7:8123
 ```
 
-2、我的`xiaozhi-esp32-server智控台`和`HomeAssistant`部署在同一台机器的另一个端口，版本是`0.3.10`
+> 手动查询 HA 的 IP 地址方法**（仅限小智esp32-server和HA部署在同一个网络设备[例如同一个wifi]下）**：
+>
+> 1. 进入 Home Assistant（前端）。
+>
+> 2. 点击左下角 **设置（Settings）** → **系统（System）** → **网络（Network）**。
+>
+> 3. 滑到最底部`Home Assistant 网址(Home Assistant website)`区域，在`本地网络(local network)`中，点击`眼睛`按钮，可以看到当前使用的 IP 地址（如 `192.168.1.10`）和网络接口。点击`复制连接(copy link)`可以直接复制。
+>
+>    ![image-20250504051716417](images/image-ha-integration-01.png)
+
+或，您已经设置了直接可以访问的Home Assistant的OAuth地址，您也可以在浏览器内直接访问
+
 ```
-http://192.168.4.7:8002
+http://homeassistant.local:8123
 ```
 
+### 2. 登录`Home Assistant`拿到开发密钥
 
-## 配置步骤
+登录`HomeAssistant`，点击`左下角头像 -> 个人`，切换`安全`导航栏，划到底部`长期访问令牌`生成api_key，并复制保存，后续的方法都需要使用这个api key且仅出现一次（小tips: 您可以保存生成的二维码图像，后续可以扫描二维码再此提取api key）。
 
-### 1. 登录`HomeAssistant`整理需要控制的设备清单
+## 方法1：小智社区共建的HA调用功能
+
+### 功能描述
+
+- 如您后续需要增加新的设备，该方法需要手动重启`xiaozhi-esp32-server服务端`以此更新设备信息**（重要**）。
+
+- 需要您确保已经在HomeAssistant中集成`Xiaomi Home`，并将米家的设备导入进`HomeAssistant`。
+
+- 需要您确保`xiaozhi-esp32-server智控台`能正常使用。
+
+- 我的`xiaozhi-esp32-server智控台`和`HomeAssistant`部署在同一台机器的另一个端口，版本是`0.3.10`
+
+  ```
+  http://192.168.4.7:8002
+  ```
+
+
+### 配置步骤
+
+#### 1. 登录`HomeAssistant`整理需要控制的设备清单
 
 登录`HomeAssistant`，点击`左下角的设置`，然后进入`设备与服务`，再点击顶部的`实体`。
 
@@ -56,12 +92,7 @@ http://192.168.4.7:8002
 这段字符，我们成为“设备清单字符”需要保存好，等一下有用。
 
 
-### 2. 登录`HomeAssistant`拿到开发密钥
+#### 2. 登录`智控台`
-
-登录`HomeAssistant`，点击`左下角个人`，切换`安全`导航栏，划到底部`长期访问令牌`生成api_key。
-
-
-### 3. 登录`智控台`
 
 使用管理员账号，登录`智控台`。点击顶部菜单`参数管理`，搜索`plugins.home_assistant.`，会有三条结果出来
 
@@ -77,7 +108,7 @@ http://192.168.4.7:8123
 编辑`plugins.home_assistant.api_key`，把你从`HomeAssistant`复制过来的密钥，粘贴进去
 
 
-### 4. 设置`意图识别`函数
+#### 3. 设置`意图识别`函数
 
 在智控台，点击顶部菜单“模型配置”，在左侧栏，找到“意图识别”，找到id为`Intent_function_call`的意图，点击编辑
 
@@ -94,18 +125,124 @@ change_role;get_weather;get_news;play_music
 change_role;get_weather;get_news;play_music;hass_get_state;hass_set_state
 ```
 
-### 6. 手动重启xiaozhi-server
+#### 4. 手动重启xiaozhi-server
 
 重启xiaozhi-server程序
 
 
-### 5. 确认角色配置是否设置了函数意图识别
+#### 5. 确认角色配置是否设置了函数意图识别
 
 在智控台，点击顶部菜单“智能体管理”，找到设备所在的智能体，点击“配置角色”
 
 确认意图识别(Intent)，是否选择“函数调用意图识别”
 
 
-### 6. 唤醒设别进行控制
+#### 6. 唤醒设别进行控制
 
-尝试和esp32说，“打开XXX灯”
+尝试和esp32说，“打开XXX灯”
+
+## 方法2：小智将Home Assistant的语音助手作为LLM工具
+
+### 功能描述
+
+- 该方法有一个比较严重的缺点——**该方法无法使用小智开源生态的function_call插件功能的能力**，因为使用Home Assistant作为小智的LLM工具会将意图识别能力转让给Home Assistant。但是**这个方法是能体验到原生的Home Assistant操作功能，且小智的聊天能力不变**。如实在介意可以使用同样是Home Assistant支持的[方法3](##方法3：使用Home Assistant的MCP服务（推荐）)，能够最大程度体验到Home Assistant的功能。
+
+### 配置步骤：
+
+#### 1. 配置Home Assistant的大模型语音助手。
+
+**需要您提前配置好Home Assistant的语音助手或大模型工具。**
+
+#### 2. 获取Home Assistant的语言助手的Agent ID.
+
+1. 进入Home Assistant页面内。左侧点击`开发者助手`。
+2. 在打开的`开发者助手`内，点击`动作`选项卡（如图示操作1），在页面内的选项栏`动作`中，找到或输入`conversation.process（对话-处理）`并选择`对话（conversation）: 处理`（如图示操作2）。
+
+![image-20250504043539343](images/image-ha-integration-02.png)
+
+3. 在页面内勾选`代理(agent)`选项，在变成常亮的`对话代理(conversation agent)`内选择您步骤一配置好的语音助手名称，如图示，我这边配置好的是`ZhipuAi`并选择。
+
+![image-20250504043854760](images/image-ha-integration-03.png)
+
+4. 选中后，点击表单左下方的`进入YAML模式`。
+
+![image-20250504043951126](images/image-ha-integration-04.png)
+
+5. 复制其中的agent-id的值，例如图示中我的是`01JP2DYMBDF7F4ZA2DMCF2AGX2`(仅供参考)。
+
+![image-20250504044046466](images/image-ha-integration-05.png)
+
+6. 切换到小智开源服务端`xiaozhi-esp32-server`的`config.yaml`文件内，在LLM配置中，找到Home Assistant，设置您的Home Assistant的网络地址，Api key和刚刚查询到的agent_id。
+7. 修改`config.yaml`文件内的`selected_module`属性的`LLM`为`HomeAssistant`，`Intent`为`nointent`。
+8. 重启小智开源服务端`xiaozhi-esp32-server`即可正常使用。
+
+## 方法3：使用Home Assistant的MCP服务（推荐）
+
+### 功能描述
+
+- 需要您提前在Home Assistant内集成并安装好HA集成——[Model Context Protocol Server](https://www.home-assistant.io/integrations/mcp_server/)。
+
+- 这个方法与方法2都是HA官方提供的解决方法，与方法2不同的是，您可以正常使用小智开源服务端`xiaozhi-esp32-server`的开源共建的插件，同时允许您随意使用任何一个支持function_call功能的LLM大模型。
+
+### 配置步骤
+
+#### 1. 安装Home Assistant的MCP服务集成。
+
+集成官方网址——[Model Context Protocol Server](https://www.home-assistant.io/integrations/mcp_server/)。。
+
+或跟随以下手动操作。
+
+> - 前往Home Assistant页面的**[设置 > 设备和服务（Settings > Devices & Services.）](https://my.home-assistant.io/redirect/integrations)**。
+>
+> - 在右下角，选择 **[添加集成（Add Integration）](https://my.home-assistant.io/redirect/config_flow_start?domain=mcp_server)**按钮。
+>
+> - 从列表中选择**模型上下文协议服务器（Model Context Protocol Server）**。
+>
+> - 按照屏幕上的说明完成设置。
+
+#### 2. 配置小智开源服务端MCP配置信息
+
+切换到小智开源服务端`xiaozhi-esp32-server`的`mcp_server_settings.json`文件内，在`"mcpServers"`的括号内添加以下内容：
+
+```json
+"Home Assistant": {
+      "command": "mcp-proxy",
+      "args": [
+        "http://YOUR_HA_HOST/mcp_server/sse"
+      ],
+      "env": {
+        "API_ACCESS_TOKEN": "YOUR_API_ACCESS_TOKEN"
+      }
+},
+```
+
+注意：
+
+1. **替换配置：**
+   - 替换`args`内的`YOUR_HA_HOST`为您的HA服务地址，如果你的服务地址已经包含了https/http字样（例如`http://192.168.1.101:8123`)，则只需要填入`192.168.1.101:8123`即可。
+   - 将`env`内`API_ACCESS_TOKEN`的`YOUR_API_ACCESS_TOKEN`替换成您之前获取到的开发密钥api key。
+2. **如果你添加配置是在`"mcpServers"`的括号内后续没有新的`mcpServers`的配置时，需要把最后的逗号`,`移除**，否则可能会解析失败。
+
+**最后效果参考以下（参考如下）**：
+
+```json
+ "mcpServers": {
+    "Home Assistant": {
+      "command": "mcp-proxy",
+      "args": [
+        "http://192.168.1.101:8123/mcp_server/sse"
+      ],
+      "env": {
+        "API_ACCESS_TOKEN": "abcd.efghi.jkl"
+      }
+    }
+  }
+```
+
+#### 3. 配置小智开源服务端的系统配置
+
+1. **选择任意一款支持function_call的LLM大模型作为小智的LLM聊天助手（但不要选择Home Assistant作为LLM工具）**，本次我选择的模型是：免费的ChatGLM，它支持functioncall函数调用，但部分时候调用不太稳定，如果像追求稳定建议把LLM设置成：DoubaoLLM，使用的具体model_name是：doubao-1-5-pro-32k-250115。
+
+2. 切换到小智开源服务端`xiaozhi-esp32-server`的`config.yaml`文件内，设置您的LLM大模型配置，并且将`selected_module`配置的`Intent`调整为`function_call`。
+
+3. 重启小智开源服务端`xiaozhi-esp32-server`即可正常使用。

main/README.md

@@ -16,4 +16,4 @@ https://ccnphfhqs21z.feishu.cn/wiki/M0XiwldO9iJwHikpXD5cEx71nKh
 
 # manager-web 、manager-api接口协议
 
-https://2662r3426b.vicp.fun/xiaozhi/v1/doc.html
+https://2662r3426b.vicp.fun/xiaozhi/doc.html

main/manager-api/src/main/java/xiaozhi/common/config/AsyncConfig.java

@@ -0,0 +1,40 @@
+package xiaozhi.common.config;
+
+import java.util.concurrent.Executor;
+import java.util.concurrent.RejectedExecutionHandler;
+import java.util.concurrent.ThreadPoolExecutor;
+
+import org.springframework.context.annotation.Bean;
+import org.springframework.context.annotation.Configuration;
+import org.springframework.context.annotation.EnableAspectJAutoProxy;
+import org.springframework.scheduling.annotation.EnableAsync;
+import org.springframework.scheduling.concurrent.ThreadPoolTaskExecutor;
+
+@Configuration
+@EnableAsync
+@EnableAspectJAutoProxy(exposeProxy = true)
+public class AsyncConfig {
+
+    @Bean(name = "taskExecutor")
+    public Executor taskExecutor() {
+        ThreadPoolTaskExecutor executor = new ThreadPoolTaskExecutor();
+        executor.setCorePoolSize(2);
+        executor.setMaxPoolSize(4);
+        executor.setQueueCapacity(1000);
+        executor.setThreadNamePrefix("AsyncThread-");
+        // 设置拒绝策略：由调用线程执行
+        executor.setRejectedExecutionHandler(new RejectedExecutionHandler() {
+            @Override
+            public void rejectedExecution(Runnable r, ThreadPoolExecutor executor) {
+                try {
+                    // 如果线程池已满，则由调用线程执行
+                    r.run();
+                } catch (Exception e) {
+                    throw new RuntimeException("执行异步任务失败", e);
+                }
+            }
+        });
+        executor.initialize();
+        return executor;
+    }
+}

main/manager-api/src/main/java/xiaozhi/common/constant/Constant.java

@@ -1,5 +1,7 @@
 package xiaozhi.common.constant;
 
+import lombok.Getter;
+
 /**
  * 常量
  * Copyright (c) 人人开源 All rights reserved.
@@ -109,6 +111,11 @@ public interface Constant {
      */
     String FILE_EXTENSION_SEG = ".";
 
+    /**
+     * 无记忆
+     */
+    String MEMORY_NO_MEM = "Memory_nomem";
+
     enum SysBaseParam {
         /**
          * 系统全称
@@ -174,8 +181,23 @@ public interface Constant {
         }
     }
 
+    @Getter
+    enum ChatHistoryConfEnum {
+        IGNORE(0, "不记录"),
+        RECORD_TEXT(1, "记录文本"),
+        RECORD_TEXT_AUDIO(2, "文本音频都记录");
+
+        private final int code;
+        private final String name;
+
+        ChatHistoryConfEnum(int code, String name) {
+            this.code = code;
+            this.name = name;
+        }
+    }
+
     /**
      * 版本号
      */
-    public static final String VERSION = "0.4.1";
+    public static final String VERSION = "0.4.3";
 }

main/manager-api/src/main/java/xiaozhi/modules/agent/controller/AgentController.java

@@ -39,6 +39,7 @@ import xiaozhi.modules.agent.dto.AgentChatHistoryDTO;
 import xiaozhi.modules.agent.dto.AgentChatSessionDTO;
 import xiaozhi.modules.agent.dto.AgentCreateDTO;
 import xiaozhi.modules.agent.dto.AgentDTO;
+import xiaozhi.modules.agent.dto.AgentMemoryDTO;
 import xiaozhi.modules.agent.dto.AgentUpdateDTO;
 import xiaozhi.modules.agent.entity.AgentEntity;
 import xiaozhi.modules.agent.entity.AgentTemplateEntity;
@@ -46,6 +47,7 @@ import xiaozhi.modules.agent.service.AgentChatAudioService;
 import xiaozhi.modules.agent.service.AgentChatHistoryService;
 import xiaozhi.modules.agent.service.AgentService;
 import xiaozhi.modules.agent.service.AgentTemplateService;
+import xiaozhi.modules.device.entity.DeviceEntity;
 import xiaozhi.modules.device.service.DeviceService;
 import xiaozhi.modules.security.user.SecurityUser;
 
@@ -109,6 +111,8 @@ public class AgentController {
             entity.setMemModelId(template.getMemModelId());
             entity.setIntentModelId(template.getIntentModelId());
             entity.setSystemPrompt(template.getSystemPrompt());
+            entity.setSummaryMemory(template.getSummaryMemory());
+            entity.setChatHistoryConf(template.getChatHistoryConf());
             entity.setLangCode(template.getLangCode());
             entity.setLanguage(template.getLanguage());
         }
@@ -125,10 +129,26 @@ public class AgentController {
         return new Result<String>().ok(entity.getId());
     }
 
+    @PutMapping("/saveMemory/{macAddress}")
+    @Operation(summary = "根据设备id更新智能体")
+    public Result<Void> updateByDeviceId(@PathVariable String macAddress, @RequestBody @Valid AgentMemoryDTO dto) {
+        DeviceEntity device = deviceService.getDeviceByMacAddress(macAddress);
+        if (device == null) {
+            return new Result<>();
+        }
+        AgentUpdateDTO agentUpdateDTO = new AgentUpdateDTO();
+        agentUpdateDTO.setSummaryMemory(dto.getSummaryMemory());
+        return updateAgentById(device.getAgentId(), agentUpdateDTO);
+    }
+
     @PutMapping("/{id}")
     @Operation(summary = "更新智能体")
     @RequiresPermissions("sys:role:normal")
     public Result<Void> update(@PathVariable String id, @RequestBody @Valid AgentUpdateDTO dto) {
+        return updateAgentById(id, dto);
+    }
+
+    private Result<Void> updateAgentById(String id, AgentUpdateDTO dto) {
         // 先查询现有实体
         AgentEntity existingEntity = agentService.getAgentById(id);
         if (existingEntity == null) {
@@ -166,6 +186,12 @@ public class AgentController {
         if (dto.getSystemPrompt() != null) {
             existingEntity.setSystemPrompt(dto.getSystemPrompt());
         }
+        if (dto.getSummaryMemory() != null) {
+            existingEntity.setSummaryMemory(dto.getSummaryMemory());
+        }
+        if (dto.getChatHistoryConf() != null) {
+            existingEntity.setChatHistoryConf(dto.getChatHistoryConf());
+        }
         if (dto.getLangCode() != null) {
             existingEntity.setLangCode(dto.getLangCode());
         }
@@ -181,8 +207,16 @@ public class AgentController {
         existingEntity.setUpdater(user.getId());
         existingEntity.setUpdatedAt(new Date());
 
+        // 更新记忆策略
+        if (existingEntity.getMemModelId() == null || existingEntity.getMemModelId().equals(Constant.MEMORY_NO_MEM)) {
+            // 删除所有记录
+            agentChatHistoryService.deleteByAgentId(existingEntity.getId(), true, true);
+            existingEntity.setSummaryMemory("");
+        } else if (existingEntity.getChatHistoryConf() != null && existingEntity.getChatHistoryConf() == 1) {
+            // 删除音频数据
+            agentChatHistoryService.deleteByAgentId(existingEntity.getId(), true, false);
+        }
         agentService.updateById(existingEntity);
-
         return new Result<>();
     }
 
@@ -193,7 +227,7 @@ public class AgentController {
         // 先删除关联的设备
         deviceService.deleteByAgentId(id);
         // 删除关联的聊天记录
-        agentChatHistoryService.deleteByAgentId(id);
+        agentChatHistoryService.deleteByAgentId(id, true, true);
         // 再删除智能体
         agentService.deleteById(id);
         return new Result<>();

main/manager-api/src/main/java/xiaozhi/modules/agent/dao/AiAgentChatHistoryDao.java

@@ -28,4 +28,11 @@ public interface AiAgentChatHistoryDao extends BaseMapper<AgentChatHistoryEntity
      * @param agentId 智能体ID
      */
     void deleteHistoryByAgentId(String agentId);
+
+    /**
+     * 根据智能体ID删除音频ID
+     *
+     * @param agentId 智能体ID
+     */
+    void deleteAudioIdByAgentId(String agentId);
 }

main/manager-api/src/main/java/xiaozhi/modules/agent/dto/AgentDTO.java

@@ -27,9 +27,16 @@ public class AgentDTO {
     @Schema(description = "大语言模型名称", example = "llm_model_01")
     private String llmModelName;
 
+    @Schema(description = "记忆模型ID", example = "mem_model_01")
+    private String memModelId;
+
     @Schema(description = "角色设定参数", example = "你是一个专业的客服助手，负责回答用户问题并提供帮助")
     private String systemPrompt;
 
+    @Schema(description = "总结记忆", example = "构建可生长的动态记忆网络，在有限空间内保留关键信息的同时，智能维护信息演变轨迹\n" +
+            "根据对话记录，总结user的重要信息，以便在未来的对话中提供更个性化的服务", required = false)
+    private String summaryMemory;
+
     @Schema(description = "最后连接时间", example = "2024-03-20 10:00:00")
     private Date lastConnectedAt;
 

main/manager-api/src/main/java/xiaozhi/modules/agent/dto/AgentMemoryDTO.java

@@ -0,0 +1,19 @@
+package xiaozhi.modules.agent.dto;
+
+import java.io.Serializable;
+
+import io.swagger.v3.oas.annotations.media.Schema;
+import lombok.Data;
+
+/**
+ * 智能体记忆更新DTO
+ */
+@Data
+@Schema(description = "智能体记忆更新对象")
+public class AgentMemoryDTO implements Serializable {
+    private static final long serialVersionUID = 1L;
+
+    @Schema(description = "总结记忆", example = "构建可生长的动态记忆网络，在有限空间内保留关键信息的同时，智能维护信息演变轨迹\n" +
+            "根据对话记录，总结user的重要信息，以便在未来的对话中提供更个性化的服务", required = false)
+    private String summaryMemory;
+}

main/manager-api/src/main/java/xiaozhi/modules/agent/dto/AgentUpdateDTO.java

@@ -45,6 +45,13 @@ public class AgentUpdateDTO implements Serializable {
     @Schema(description = "角色设定参数", example = "你是一个专业的客服助手，负责回答用户问题并提供帮助", required = false)
     private String systemPrompt;
 
+    @Schema(description = "总结记忆", example = "构建可生长的动态记忆网络，在有限空间内保留关键信息的同时，智能维护信息演变轨迹\n" +
+            "根据对话记录，总结user的重要信息，以便在未来的对话中提供更个性化的服务", required = false)
+    private String summaryMemory;
+
+    @Schema(description = "聊天记录配置（0不记录 1仅记录文本 2记录文本和语音）", example = "3", required = false)
+    private Integer chatHistoryConf;
+
     @Schema(description = "语言编码", example = "zh_CN", required = false)
     private String langCode;
 

main/manager-api/src/main/java/xiaozhi/modules/agent/entity/AgentEntity.java

@@ -48,9 +48,16 @@ public class AgentEntity {
     @Schema(description = "意图模型标识")
     private String intentModelId;
 
+    @Schema(description = "聊天记录配置（0不记录 1仅记录文本 2记录文本和语音）")
+    private Integer chatHistoryConf;
+
     @Schema(description = "角色设定参数")
     private String systemPrompt;
 
+    @Schema(description = "总结记忆", example = "构建可生长的动态记忆网络，在有限空间内保留关键信息的同时，智能维护信息演变轨迹\n" +
+            "根据对话记录，总结user的重要信息，以便在未来的对话中提供更个性化的服务", required = false)
+    private String summaryMemory;
+
     @Schema(description = "语言编码")
     private String langCode;
 

main/manager-api/src/main/java/xiaozhi/modules/agent/entity/AgentTemplateEntity.java

@@ -69,11 +69,20 @@ public class AgentTemplateEntity implements Serializable {
      */
     private String intentModelId;
 
+    /**
+     * 聊天记录配置（0不记录 1仅记录文本 2记录文本和语音）
+     */
+    private Integer chatHistoryConf;
+
     /**
      * 角色设定参数
      */
     private String systemPrompt;
 
+    /**
+     * 总结记忆
+     */
+    private String summaryMemory;
     /**
      * 语言编码
      */

main/manager-api/src/main/java/xiaozhi/modules/agent/service/AgentChatHistoryService.java

@@ -39,7 +39,9 @@ public interface AgentChatHistoryService extends IService<AgentChatHistoryEntity
     /**
      * 根据智能体ID删除聊天记录
      *
-     * @param agentId 智能体ID
+     * @param agentId     智能体ID
+     * @param deleteAudio 是否删除音频
+     * @param deleteText  是否删除文本
      */
-    void deleteByAgentId(String agentId);
+    void deleteByAgentId(String agentId, Boolean deleteAudio, Boolean deleteText);
 }

main/manager-api/src/main/java/xiaozhi/modules/agent/service/biz/impl/AgentChatHistoryBizServiceImpl.java

@@ -2,12 +2,14 @@ package xiaozhi.modules.agent.service.biz.impl;
 
 import java.util.Base64;
 import java.util.Date;
+import java.util.Objects;
 
 import org.springframework.stereotype.Service;
 import org.springframework.transaction.annotation.Transactional;
 
 import lombok.RequiredArgsConstructor;
 import lombok.extern.slf4j.Slf4j;
+import xiaozhi.common.constant.Constant;
 import xiaozhi.common.redis.RedisKeys;
 import xiaozhi.common.redis.RedisUtils;
 import xiaozhi.modules.agent.dto.AgentChatHistoryReportDTO;
@@ -47,8 +49,33 @@ public class AgentChatHistoryBizServiceImpl implements AgentChatHistoryBizServic
         Byte chatType = report.getChatType();
         log.info("小智设备聊天上报请求: macAddress={}, type={}", macAddress, chatType);
 
-        // 1. base64解码report.getOpusDataBase64(),存入ai_agent_chat_audio表
+        // 根据设备MAC地址查询对应的默认智能体，判断是否需要上报
+        AgentEntity agentEntity = agentService.getDefaultAgentByMacAddress(macAddress);
+        if (agentEntity == null) {
+            return Boolean.FALSE;
+        }
+
+        Integer chatHistoryConf = agentEntity.getChatHistoryConf();
+        String agentId = agentEntity.getId();
+
+        if (Objects.equals(chatHistoryConf, Constant.ChatHistoryConfEnum.RECORD_TEXT.getCode())) {
+            saveChatText(report, agentId, macAddress, null);
+        } else if (Objects.equals(chatHistoryConf, Constant.ChatHistoryConfEnum.RECORD_TEXT_AUDIO.getCode())) {
+            String audioId = saveChatAudio(report);
+            saveChatText(report, agentId, macAddress, audioId);
+        }
+
+        // 更新设备最后对话时间
+        redisUtils.set(RedisKeys.getAgentDeviceLastConnectedAtById(agentId), new Date());
+        return Boolean.TRUE;
+    }
+
+    /**
+     * base64解码report.getOpusDataBase64(),存入ai_agent_chat_audio表
+     */
+    private String saveChatAudio(AgentChatHistoryReportDTO report) {
         String audioId = null;
+
         if (report.getAudioBase64() != null && !report.getAudioBase64().isEmpty()) {
             try {
                 byte[] audioData = Base64.getDecoder().decode(report.getAudioBase64());
@@ -56,20 +83,18 @@ public class AgentChatHistoryBizServiceImpl implements AgentChatHistoryBizServic
                 log.info("音频数据保存成功，audioId={}", audioId);
             } catch (Exception e) {
                 log.error("音频数据保存失败", e);
-                return false;
+                return null;
             }
         }
+        return audioId;
+    }
 
-        // 2. 组装上报数据
+    /**
-        // 2.1 根据设备MAC地址查询对应的默认智能体，判断是否需要上报
+     * 组装上报数据
-        AgentEntity agentEntity = agentService.getDefaultAgentByMacAddress(macAddress);
+     */
-        if (agentEntity == null) {
+    private void saveChatText(AgentChatHistoryReportDTO report, String agentId, String macAddress, String audioId) {
-            return false;
-        }
-        String agentId = agentEntity.getId();
-        log.info("设备 {} 对应智能体 {} 上报", macAddress, agentEntity.getId());
 
-        // 2.2 构建聊天记录实体
+        // 构建聊天记录实体
         AgentChatHistoryEntity entity = AgentChatHistoryEntity.builder()
                 .macAddress(macAddress)
                 .agentId(agentId)
@@ -79,10 +104,9 @@ public class AgentChatHistoryBizServiceImpl implements AgentChatHistoryBizServic
                 .audioId(audioId)
                 .build();
 
-        // 3. 保存数据
+        // 保存数据
         agentChatHistoryService.save(entity);
-        // 4. 更新设备最后对话时间
+
-        redisUtils.set(RedisKeys.getAgentDeviceLastConnectedAtById(agentId), new Date());
+        log.info("设备 {} 对应智能体 {} 上报成功", macAddress, agentId);
-        return Boolean.TRUE;
     }
 }

main/manager-api/src/main/java/xiaozhi/modules/agent/service/impl/AgentChatHistoryServiceImpl.java

@@ -78,8 +78,16 @@ public class AgentChatHistoryServiceImpl extends ServiceImpl<AiAgentChatHistoryD
 
     @Override
     @Transactional(rollbackFor = Exception.class)
-    public void deleteByAgentId(String agentId) {
+    public void deleteByAgentId(String agentId, Boolean deleteAudio, Boolean deleteText) {
-        baseMapper.deleteAudioByAgentId(agentId);
+        if (deleteAudio) {
-        baseMapper.deleteHistoryByAgentId(agentId);
+            baseMapper.deleteAudioByAgentId(agentId);
+        }
+        if (deleteAudio && !deleteText) {
+            baseMapper.deleteAudioIdByAgentId(agentId);
+        }
+        if (deleteText) {
+            baseMapper.deleteHistoryByAgentId(agentId);
+        }
+
     }
 }

main/manager-api/src/main/java/xiaozhi/modules/agent/service/impl/AgentServiceImpl.java

@@ -13,6 +13,7 @@ import com.baomidou.mybatisplus.core.conditions.update.UpdateWrapper;
 import com.baomidou.mybatisplus.core.metadata.IPage;
 
 import lombok.AllArgsConstructor;
+import xiaozhi.common.constant.Constant;
 import xiaozhi.common.page.PageData;
 import xiaozhi.common.redis.RedisKeys;
 import xiaozhi.common.redis.RedisUtils;
@@ -46,7 +47,15 @@ public class AgentServiceImpl extends BaseServiceImpl<AgentDao, AgentEntity> imp
 
     @Override
     public AgentEntity getAgentById(String id) {
-        return agentDao.selectById(id);
+        AgentEntity agent = agentDao.selectById(id);
+        if (agent != null && agent.getMemModelId() != null && agent.getMemModelId().equals(Constant.MEMORY_NO_MEM)) {
+            agent.setChatHistoryConf(Constant.ChatHistoryConfEnum.IGNORE.getCode());
+        } else if (agent != null && agent.getMemModelId() != null
+                && !agent.getMemModelId().equals(Constant.MEMORY_NO_MEM)
+                && agent.getChatHistoryConf() == null) {
+            agent.setChatHistoryConf(Constant.ChatHistoryConfEnum.RECORD_TEXT_AUDIO.getCode());
+        }
+        return agent;
     }
 
     @Override
@@ -93,6 +102,9 @@ public class AgentServiceImpl extends BaseServiceImpl<AgentDao, AgentEntity> imp
             // 获取 LLM 模型名称
             dto.setLlmModelName(modelConfigService.getModelNameById(agent.getLlmModelId()));
 
+            // 获取记忆模型名称
+            dto.setMemModelId(agent.getMemModelId());
+
             // 获取 TTS 音色名称
             dto.setTtsVoiceName(timbreModelService.getTimbreNameById(agent.getTtsVoiceId()));
 

main/manager-api/src/main/java/xiaozhi/modules/config/service/impl/ConfigServiceImpl.java

@@ -9,6 +9,7 @@ import org.apache.commons.lang3.StringUtils;
 import org.springframework.stereotype.Service;
 
 import lombok.AllArgsConstructor;
+import xiaozhi.common.constant.Constant;
 import xiaozhi.common.exception.ErrorCode;
 import xiaozhi.common.exception.RenException;
 import xiaozhi.common.redis.RedisKeys;
@@ -64,6 +65,7 @@ public class ConfigServiceImpl implements ConfigService {
                 null,
                 null,
                 null,
+                null,
                 agent.getVadModelId(),
                 agent.getAsrModelId(),
                 null,
@@ -108,6 +110,17 @@ public class ConfigServiceImpl implements ConfigService {
         // 获取单台设备每天最多输出字数
         String deviceMaxOutputSize = sysParamsService.getValue("device_max_output_size", true);
         result.put("device_max_output_size", deviceMaxOutputSize);
+
+        // 获取聊天记录配置
+        Integer chatHistoryConf = agent.getChatHistoryConf();
+        if (agent.getMemModelId() != null && agent.getMemModelId().equals(Constant.MEMORY_NO_MEM)) {
+            chatHistoryConf = Constant.ChatHistoryConfEnum.IGNORE.getCode();
+        } else if (agent.getMemModelId() != null
+                && !agent.getMemModelId().equals(Constant.MEMORY_NO_MEM)
+                && agent.getChatHistoryConf() == null) {
+            chatHistoryConf = Constant.ChatHistoryConfEnum.RECORD_TEXT_AUDIO.getCode();
+        }
+        result.put("chat_history_conf", chatHistoryConf);
         // 如果客户端已实例化模型，则不返回
         String alreadySelectedVadModelId = (String) selectedModule.get("VAD");
         if (alreadySelectedVadModelId != null && alreadySelectedVadModelId.equals(agent.getVadModelId())) {
@@ -122,6 +135,7 @@ public class ConfigServiceImpl implements ConfigService {
         buildModuleConfig(
                 agent.getAgentName(),
                 agent.getSystemPrompt(),
+                agent.getSummaryMemory(),
                 voice,
                 agent.getVadModelId(),
                 agent.getAsrModelId(),
@@ -222,6 +236,7 @@ public class ConfigServiceImpl implements ConfigService {
     private void buildModuleConfig(
             String assistantName,
             String prompt,
+            String summaryMemory,
             String voice,
             String vadModelId,
             String asrModelId,
@@ -257,7 +272,8 @@ public class ConfigServiceImpl implements ConfigService {
                         if (intentLLMModelId != null && intentLLMModelId.equals(llmModelId)) {
                             intentLLMModelId = null;
                         }
-                    } else if ("function_call".equals(map.get("type"))) {
+                    }
+                    if (map.get("functions") != null) {
                         String functionStr = (String) map.get("functions");
                         if (StringUtils.isNotBlank(functionStr)) {
                             String[] functions = functionStr.split("\\;");
@@ -281,5 +297,6 @@ public class ConfigServiceImpl implements ConfigService {
             prompt = prompt.replace("{{assistant_name}}", StringUtils.isBlank(assistantName) ? "小智" : assistantName);
         }
         result.put("prompt", prompt);
+        result.put("summaryMemory", summaryMemory);
     }
 }

main/manager-api/src/main/java/xiaozhi/modules/device/dao/DeviceDao.java

@@ -1,18 +1,18 @@
 package xiaozhi.modules.device.dao;
 
+import java.util.Date;
+
 import org.apache.ibatis.annotations.Mapper;
 
 import com.baomidou.mybatisplus.core.mapper.BaseMapper;
 
 import xiaozhi.modules.device.entity.DeviceEntity;
 
-import java.util.Date;
-import java.util.List;
-
 @Mapper
 public interface DeviceDao extends BaseMapper<DeviceEntity> {
     /**
      * 获取此智能体全部设备的最后连接时间
+     * 
      * @param agentId 智能体id
      * @return
      */

main/manager-api/src/main/java/xiaozhi/modules/device/service/impl/DeviceServiceImpl.java

@@ -9,6 +9,8 @@ import java.util.TimeZone;
 import java.util.UUID;
 
 import org.apache.commons.lang3.StringUtils;
+import org.springframework.aop.framework.AopContext;
+import org.springframework.scheduling.annotation.Async;
 import org.springframework.stereotype.Service;
 import org.springframework.web.context.request.RequestContextHolder;
 import org.springframework.web.context.request.ServletRequestAttributes;
@@ -54,6 +56,24 @@ public class DeviceServiceImpl extends BaseServiceImpl<DeviceDao, DeviceEntity>
     private final RedisUtils redisUtils;
     private final OtaService otaService;
 
+    @Async
+    public void updateDeviceConnectionInfo(String agentId, String deviceId, String appVersion) {
+        try {
+            DeviceEntity device = new DeviceEntity();
+            device.setId(deviceId);
+            device.setLastConnectedAt(new Date());
+            if (StringUtils.isNotBlank(appVersion)) {
+                device.setAppVersion(appVersion);
+            }
+            deviceDao.updateById(device);
+            if (StringUtils.isNotBlank(agentId)) {
+                redisUtils.set(RedisKeys.getAgentDeviceLastConnectedAtById(agentId), new Date());
+            }
+        } catch (Exception e) {
+            log.error("异步更新设备连接信息失败", e);
+        }
+    }
+
     @Override
     public Boolean deviceActivation(String agentId, String activationCode) {
         if (StringUtils.isBlank(activationCode)) {
@@ -118,13 +138,20 @@ public class DeviceServiceImpl extends BaseServiceImpl<DeviceDao, DeviceEntity>
 
         DeviceEntity deviceById = getDeviceByMacAddress(macAddress);
 
-        if (deviceById == null || deviceById.getAutoUpdate() != 0) {
+        // 设备未绑定，则返回当前上传的固件信息（不更新）以此兼容旧固件版本
-            String type = deviceReport.getBoard() == null ? null : deviceReport.getBoard().getType();
+        if (deviceById == null) {
-            DeviceReportRespDTO.Firmware firmware = buildFirmwareInfo(type,
+            DeviceReportRespDTO.Firmware firmware = new DeviceReportRespDTO.Firmware();
-                    deviceReport.getApplication() == null ? null : deviceReport.getApplication().getVersion());
+            firmware.setVersion(deviceReport.getApplication().getVersion());
+            firmware.setUrl("http://xiaozhi.server.com:8002/xiaozhi/otaMag/download/NOT_ACTIVATED_FIRMWARE_THIS_IS_A_INVALID_URL");
             response.setFirmware(firmware);
         } else {
-            response.setFirmware(null);
+            // 只有在设备已绑定且autoUpdate不为0的情况下才返回固件升级信息
+            if (deviceById.getAutoUpdate() != 0) {
+                String type = deviceReport.getBoard() == null ? null : deviceReport.getBoard().getType();
+                DeviceReportRespDTO.Firmware firmware = buildFirmwareInfo(type,
+                        deviceReport.getApplication() == null ? null : deviceReport.getApplication().getVersion());
+                response.setFirmware(firmware);
+            }
         }
 
         // 添加WebSocket配置
@@ -149,13 +176,12 @@ public class DeviceServiceImpl extends BaseServiceImpl<DeviceDao, DeviceEntity>
         response.setWebsocket(websocket);
 
         if (deviceById != null) {
-            // 如果设备存在，则更新上次连接时间
+            // 如果设备存在，则异步更新上次连接时间和版本信息
-            deviceById.setLastConnectedAt(new Date());
+            String appVersion = deviceReport.getApplication() != null ? deviceReport.getApplication().getVersion()
-            if (deviceReport.getApplication() != null
+                    : null;
-                    && StringUtils.isNotBlank(deviceReport.getApplication().getVersion())) {
+            // 通过Spring代理调用异步方法
-                deviceById.setAppVersion(deviceReport.getApplication().getVersion());
+            ((DeviceServiceImpl) AopContext.currentProxy()).updateDeviceConnectionInfo(deviceById.getAgentId(),
-            }
+                    deviceById.getId(), appVersion);
-            deviceDao.updateById(deviceById);
         } else {
             // 如果设备不存在，则生成激活码
             DeviceReportRespDTO.Activation code = buildActivation(macAddress, deviceReport);
@@ -384,4 +410,4 @@ public class DeviceServiceImpl extends BaseServiceImpl<DeviceDao, DeviceEntity>
         }
         return 0;
     }
-}
+}

main/manager-api/src/main/java/xiaozhi/modules/security/config/ShiroConfig.java

@@ -86,6 +86,7 @@ public class ShiroConfig {
         // 将config路径使用server服务过滤器
         filterMap.put("/config/**", "server");
         filterMap.put("/agent/chat-history/report", "server");
+        filterMap.put("/agent/saveMemory/**", "server");
         filterMap.put("/agent/play/**", "anon");
         filterMap.put("/**", "oauth2");
         shiroFilter.setFilterChainDefinitionMap(filterMap);

main/manager-api/src/main/java/xiaozhi/modules/sys/dto/SysParamsDTO.java

@@ -39,7 +39,7 @@ public class SysParamsDTO implements Serializable {
 
     @Schema(description = "值类型")
     @NotBlank(message = "{sysparams.valuetype.require}", groups = DefaultGroup.class)
-    @Pattern(regexp = "^(string|number|boolean|array)$", message = "{sysparams.valuetype.pattern}", groups = DefaultGroup.class)
+    @Pattern(regexp = "^(string|number|boolean|array|json)$", message = "{sysparams.valuetype.pattern}", groups = DefaultGroup.class)
     private String valueType;
 
     @Schema(description = "备注")

main/manager-api/src/main/java/xiaozhi/modules/sys/service/impl/SysParamsServiceImpl.java

@@ -127,6 +127,12 @@ public class SysParamsServiceImpl extends BaseServiceImpl<SysParamsDao, SysParam
                 break;
             case "json":
                 try {
+                    // 首先检查是否以 { 开头，以 } 结尾
+                    String trimmedValue = paramValue.trim();
+                    if (!trimmedValue.startsWith("{") || !trimmedValue.endsWith("}")) {
+                        throw new RenException(ErrorCode.PARAM_JSON_INVALID);
+                    }
+                    // 然后尝试解析JSON
                     JsonUtils.parseObject(paramValue, Object.class);
                 } catch (Exception e) {
                     throw new RenException(ErrorCode.PARAM_JSON_INVALID);

main/manager-api/src/main/resources/db/changelog/202505081146.sql

@@ -0,0 +1,43 @@
+-- 添加百度ASR模型配置
+delete from `ai_model_config` where `id` = 'ASR_BaiduASR';
+INSERT INTO `ai_model_config` VALUES ('ASR_BaiduASR', 'ASR', 'BaiduASR', '百度语音识别', 0, 1, '{\"type\": \"baidu\", \"app_id\": \"\", \"api_key\": \"\", \"secret_key\": \"\", \"dev_pid\": 1537, \"output_dir\": \"tmp/\"}', NULL, NULL, 7, NULL, NULL, NULL, NULL);
+
+
+-- 添加百度ASR供应器
+delete from `ai_model_provider` where `id` = 'SYSTEM_ASR_BaiduASR';
+INSERT INTO `ai_model_provider` (`id`, `model_type`, `provider_code`, `name`, `fields`, `sort`, `creator`, `create_date`, `updater`, `update_date`) VALUES
+('SYSTEM_ASR_BaiduASR', 'ASR', 'baidu', '百度语音识别', '[{"key":"app_id","label":"应用AppID","type":"string"},{"key":"api_key","label":"API Key","type":"string"},{"key":"secret_key","label":"Secret Key","type":"string"},{"key":"dev_pid","label":"语言参数","type":"number"},{"key":"output_dir","label":"输出目录","type":"string"}]', 7, 1, NOW(), 1, NOW());
+
+
+-- 更新百度ASR配置说明
+UPDATE `ai_model_config` SET 
+`doc_link` = 'https://console.bce.baidu.com/ai-engine/old/#/ai/speech/app/list',
+`remark` = '百度ASR配置说明：
+1. 访问 https://console.bce.baidu.com/ai-engine/old/#/ai/speech/app/list
+2. 创建新应用
+3. 获取AppID、API Key和Secret Key
+4. 填入配置文件中
+查看资源额度：https://console.bce.baidu.com/ai-engine/old/#/ai/speech/overview/resource/list
+语言参数说明：https://ai.baidu.com/ai-doc/SPEECH/0lbxfnc9b
+' WHERE `id` = 'ASR_BaiduASR';
+
+-- 更新豆包供应器字段
+update `ai_model_provider` set `fields` = 
+'[{"key":"appid","label":"应用ID","type":"string"},{"key":"access_token","label":"访问令牌","type":"string"},{"key":"cluster","label":"集群","type":"string"},{"key":"boosting_table_name","label":"热词文件名称","type":"string"},{"key":"correct_table_name","label":"替换词文件名称","type":"string"},{"key":"output_dir","label":"输出目录","type":"string"}]'
+where `id` = 'SYSTEM_ASR_DoubaoASR';
+
+-- 更新豆包ASR配置说明
+UPDATE `ai_model_config` SET 
+`doc_link` = 'https://console.volcengine.com/speech/app',
+`remark` = '豆包ASR配置说明：
+1. 需要在火山引擎控制台创建应用并获取appid和access_token
+2. 支持中文语音识别
+3. 需要网络连接
+4. 输出文件保存在tmp/目录
+申请步骤：
+1. 访问 https://console.volcengine.com/speech/app
+2. 创建新应用
+3. 获取appid和access_token
+4. 填入配置文件中
+如需设置热词，请参考：https://www.volcengine.com/docs/6561/155738
+' WHERE `id` = 'ASR_DoubaoASR';

main/manager-api/src/main/resources/db/changelog/202505091409.sql

@@ -0,0 +1,6 @@
+-- 更新intent_llmM供应器
+update `ai_model_provider` set fields =  '[{"key":"llm","label":"LLM模型","type":"string"},{"key":"functions","label":"函数列表","type":"dict","dict_name":"functions"}]' where  id = 'SYSTEM_Intent_intent_llm';
+-- 更新ChatGLMLLM的意图识别配置
+update `ai_model_config` set config_json =  '{\"type\": \"intent_llm\", \"llm\": \"LLM_ChatGLMLLM\", \"functions\": \"get_weather;get_news_from_newsnow;play_music\"}' where  id = 'Intent_intent_llm';
+-- 更新函数调用意图识别配置
+UPDATE `ai_model_config` SET config_json = REPLACE(config_json, ';get_news;', ';get_news_from_newsnow;') WHERE id = 'Intent_function_call';

main/manager-api/src/main/resources/db/changelog/202505091555.sql

@@ -0,0 +1,25 @@
+-- 更新模型供应器表
+UPDATE `ai_model_provider` SET fields = '[{"key": "host", "type": "string", "label": "服务地址"}, {"key": "port", "type": "number", "label": "端口号"}, {"key": "type", "type": "string", "label": "服务类型"}, {"key": "is_ssl", "type": "boolean", "label": "是否使用SSL"}, {"key": "api_key", "type": "string", "label": "API密钥"}, {"key": "output_dir", "type": "string", "label": "输出目录"}]' WHERE id = 'SYSTEM_ASR_FunASRServer';
+
+-- 更新模型配置表
+UPDATE `ai_model_config` SET 
+config_json = '{"host": "127.0.0.1", "port": 10096, "type": "fun_server", "is_ssl": true, "api_key": "none", "output_dir": "tmp/"}',
+`doc_link` = 'https://github.com/modelscope/FunASR/blob/main/runtime/docs/SDK_advanced_guide_online_zh.md',
+`remark` = '独立部署FunASR，使用FunASR的API服务，只需要五句话
+第一句：mkdir -p ./funasr-runtime-resources/models
+第二句：sudo docker run -p 10096:10095 -it --privileged=true -v $PWD/funasr-runtime-resources/models:/workspace/models registry.cn-hangzhou.aliyuncs.com/funasr_repo/funasr:funasr-runtime-sdk-online-cpu-0.1.12
+上一句话执行后会进入到容器，继续第三句：cd FunASR/runtime
+不要退出容器，继续在容器中执行第四句：nohup bash run_server_2pass.sh --download-model-dir /workspace/models --vad-dir damo/speech_fsmn_vad_zh-cn-16k-common-onnx --model-dir damo/speech_paraformer-large-vad-punc_asr_nat-zh-cn-16k-common-vocab8404-onnx  --online-model-dir damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-online-onnx  --punc-dir damo/punc_ct-transformer_zh-cn-common-vad_realtime-vocab272727-onnx --lm-dir damo/speech_ngram_lm_zh-cn-ai-wesp-fst --itn-dir thuduj12/fst_itn_zh --hotword /workspace/models/hotwords.txt > log.txt 2>&1 &
+上一句话执行后会进入到容器，继续第五句：tail -f log.txt
+第五句话执行完后，会看到模型下载日志，下载完后就可以连接使用了
+以上是使用CPU推理，如果有GPU，详细参考：https://github.com/modelscope/FunASR/blob/main/runtime/docs/SDK_advanced_guide_online_zh.md' WHERE `id` = 'ASR_FunASRServer';
+
+-- FishSpeech配置说明
+UPDATE `ai_model_config` SET 
+`doc_link` = 'https://github.com/xinnan-tech/xiaozhi-esp32-server/blob/main/docs/fish-speech-integration.md',
+`remark` = 'FishSpeech配置说明：
+1. 需要本地部署FishSpeech服务
+2. 支持自定义音色
+3. 本地推理，无需网络连接
+4. 输出文件保存在tmp/目录
+5. 可参照教程https://github.com/xinnan-tech/xiaozhi-esp32-server/blob/main/docs/fish-speech-integration.md' WHERE `id` = 'TTS_FishSpeech';

main/manager-api/src/main/resources/db/changelog/202505111914.sql

@@ -0,0 +1,6 @@
+-- 添加聊天记录配置字段
+ALTER TABLE `ai_agent` 
+ADD COLUMN `chat_history_conf` tinyint NOT NULL DEFAULT 0 COMMENT '聊天记录配置（0不记录 1仅记录文本 2记录文本和语音）' AFTER `system_prompt`;
+
+ALTER TABLE `ai_agent_template` 
+ADD COLUMN `chat_history_conf` tinyint NOT NULL DEFAULT 0 COMMENT '聊天记录配置（0不记录 1仅记录文本 2记录文本和语音）' AFTER `system_prompt`;

main/manager-api/src/main/resources/db/changelog/202505122348.sql

@@ -0,0 +1,6 @@
+-- 添加总结记忆字段
+ALTER TABLE `ai_agent`
+ADD COLUMN `summary_memory` text COMMENT '总结记忆' AFTER `system_prompt`;
+
+ALTER TABLE `ai_agent_template`
+ADD COLUMN `summary_memory` text COMMENT '总结记忆' AFTER `system_prompt`;

main/manager-api/src/main/resources/db/changelog/202505142037.sql

@@ -0,0 +1,7 @@
+update ai_agent_template set system_prompt = replace(system_prompt, '我是', '你是');
+
+delete from sys_params where id in (500,501,402);
+INSERT INTO `sys_params` (id, param_code, param_value, value_type, param_type, remark) VALUES (500, 'end_prompt.enable', 'true', 'boolean', 1, '是否开启结束语');
+INSERT INTO `sys_params` (id, param_code, param_value, value_type, param_type, remark) VALUES (501, 'end_prompt.prompt', '请你以“时间过得真快”未来头，用富有感情、依依不舍的话来结束这场对话吧！', 'string', 1, '结束提示词');
+
+INSERT INTO `sys_params` (id, param_code, param_value, value_type, param_type, remark) VALUES (402, 'plugins.get_weather.api_host', 'mj7p3y7naa.re.qweatherapi.com', 'string', 1, '开发者apihost');

main/manager-api/src/main/resources/db/changelog/db.changelog-master.yaml

@@ -99,4 +99,46 @@ databaseChangeLog:
       changes:
         - sqlFile:
             encoding: utf8
-            path: classpath:db/changelog/202505022134.sql
+            path: classpath:db/changelog/202505022134.sql
+  - changeSet:
+      id: 202505081146
+      author: hrz
+      changes:
+        - sqlFile:
+            encoding: utf8
+            path: classpath:db/changelog/202505081146.sql
+  - changeSet:
+      id: 202505091409
+      author: hrz
+      changes:
+        - sqlFile:
+            encoding: utf8
+            path: classpath:db/changelog/202505091409.sql
+  - changeSet:
+      id: 202505091555
+      author: whosmyqueen
+      changes:
+        - sqlFile:
+            encoding: utf8
+            path: classpath:db/changelog/202505091555.sql
+  - changeSet:
+      id: 202505111914
+      author: hrz
+      changes:
+        - sqlFile:
+            encoding: utf8
+            path: classpath:db/changelog/202505111914.sql
+  - changeSet:
+      id: 202505122348
+      author: ljwwd2
+      changes:
+        - sqlFile:
+            encoding: utf8
+            path: classpath:db/changelog/202505122348.sql
+  - changeSet:
+      id: 202505142037
+      author: hrz
+      changes:
+        - sqlFile:
+            encoding: utf8
+            path: classpath:db/changelog/202505142037.sql

main/manager-api/src/main/resources/mapper/agent/AiAgentChatHistoryDao.xml

@@ -28,11 +28,17 @@
       SELECT audio_id 
       FROM ai_agent_chat_history 
       WHERE agent_id = #{agentId}
-    );
+    )
   </delete>
+  
+  <update id="deleteAudioIdByAgentId">
+    UPDATE ai_agent_chat_history
+    SET audio_id = NULL
+    WHERE agent_id = #{agentId}
+  </update>
 
   <delete id="deleteHistoryByAgentId">
     DELETE FROM ai_agent_chat_history 
-    WHERE agent_id = #{agentId};
+    WHERE agent_id = #{agentId}
   </delete>
 </mapper>

main/manager-web/src/apis/module/agent.js

@@ -50,9 +50,9 @@ export default {
             }).send();
     },
     // 获取智能体配置
-    getDeviceConfig(deviceId, callback) {
+    getDeviceConfig(agentId, callback) {
         RequestService.sendRequest()
-            .url(`${getServiceUrl()}/agent/${deviceId}`)
+            .url(`${getServiceUrl()}/agent/${agentId}`)
             .method('GET')
             .success((res) => {
                 RequestService.clearRequestTime();
@@ -61,7 +61,7 @@ export default {
             .fail((err) => {
                 console.error('获取配置失败:', err);
                 RequestService.reAjaxFun(() => {
-                    this.getDeviceConfig(deviceId, callback);
+                    this.getDeviceConfig(agentId, callback);
                 });
             }).send();
     },

main/manager-web/src/components/AddDeviceDialog.vue

@@ -1,5 +1,5 @@
 <template>
-  <el-dialog :visible="visible" @close="handleClose" width="400px" center>
+  <el-dialog :visible="visible" @close="handleClose" width="24%" center>
     <div
       style="margin: 0 10px 10px;display: flex;align-items: center;gap: 10px;font-weight: 700;font-size: 20px;text-align: left;color: #3d4566;">
       <div

main/manager-web/src/components/AddModelDialog.vue

@@ -1,5 +1,5 @@
 <template>
-  <el-dialog :visible="dialogVisible" @update:visible="handleVisibleChange" width="975px" center
+  <el-dialog :visible="dialogVisible" @update:visible="handleVisibleChange" width="57%" center
     custom-class="custom-dialog" :show-close="false" class="center-dialog">
     <div style="margin: 0 18px; text-align: left; padding: 10px; border-radius: 10px;">
       <div style="font-size: 30px; color: #3d4566; margin-top: -10px; margin-bottom: 10px; text-align: center;">
@@ -54,7 +54,7 @@
         </el-form-item>
 
         <el-form-item label="备注" prop="remark" class="prop-remark">
-          <el-input v-model="formData.remark" type="textarea" :rows="3" placeholder="请输入模型备注"
+          <el-input v-model="formData.remark" type="textarea" :rows="3" placeholder="请输入模型备注" :autosize="{ minRows: 3, maxRows: 5 }"
             class="custom-input-bg"></el-input>
         </el-form-item>
       </el-form>
@@ -271,7 +271,7 @@ export default {
 }
 
 .center-dialog .el-dialog {
-  margin: 4% 0 auto !important;
+  margin: 0 0 auto !important;
   display: flex;
   flex-direction: column;
 }

main/manager-web/src/components/AddWisdomBodyDialog.vue

@@ -1,5 +1,5 @@
 <template>
-  <el-dialog :visible="visible" @close="handleClose"  width="400px" center @open="handleOpen">
+  <el-dialog :visible="visible" @close="handleClose"  width="25%" center @open="handleOpen">
     <div
       style="margin: 0 10px 10px;display: flex;align-items: center;gap: 10px;font-weight: 700;font-size: 20px;text-align: left;color: #3d4566;">
       <div
@@ -10,7 +10,7 @@
     </div>
     <div style="height: 1px;background: #e8f0ff;" />
     <div style="margin: 22px 15px;">
-      <div style="font-weight: 400;font-size: 14px;text-align: left;color: #3d4566;">
+      <div style="font-weight: 400;text-align: left;color: #3d4566;">
         <div style="color: red;display: inline-block;">*</div> 智能体名称：
       </div>
       <div class="input-46" style="margin-top: 12px;">

main/manager-web/src/components/ChangePasswordDialog.vue

@@ -1,6 +1,6 @@
 <template>
   <form>
-    <el-dialog :visible.sync="value" width="400px" center>
+    <el-dialog :visible.sync="dialogVisible"  width="24%" center>
       <div
         style="margin: 0 10px 10px;display: flex;align-items: center;gap: 10px;font-weight: 700;font-size: 20px;text-align: left;color: #3d4566;">
         <div
@@ -60,11 +60,20 @@ export default {
   },
   data() {
     return {
+      dialogVisible: this.value,
       oldPassword: "",
       newPassword: "",
       confirmNewPassword: ""
     }
   },
+  watch: {
+    value(val) {
+      this.dialogVisible = val;
+    },
+    dialogVisible(val) {
+      this.$emit('input', val);
+    }
+  },
   methods: {
     ...mapActions(['logout']), // 引入Vuex的logout action
     confirm() {
@@ -101,7 +110,7 @@ export default {
       this.$emit('input', false);
     },
     cancel() {
-      this.$emit('input', false);
+      this.dialogVisible = false;
       this.resetForm();
     },
     resetForm() {

main/manager-web/src/components/DeviceItem.vue

@@ -26,8 +26,12 @@
       <div class="settings-btn" @click="handleDeviceManage">
         设备管理({{ device.deviceCount }})
       </div>
-      <div class="settings-btn" @click="handleChatHistory">
+      <div class="settings-btn" @click="handleChatHistory"
-        聊天记录
+        :class="{ 'disabled-btn': device.memModelId === 'Memory_nomem' }">
+        <el-tooltip v-if="device.memModelId === 'Memory_nomem'" content="未开启记忆" placement="top">
+          <span>聊天记录</span>
+        </el-tooltip>
+        <span v-else>聊天记录</span>
       </div>
     </div>
     <div class="version-info">
@@ -77,6 +81,9 @@ export default {
       this.$router.push({ path: '/device-management', query: { agentId: this.device.agentId } });
     },
     handleChatHistory() {
+      if (this.device.memModelId === 'Memory_nomem') {
+        return
+      }
       this.$emit('chat-history', { agentId: this.device.agentId, agentName: this.device.agentName })
     }
   }
@@ -120,6 +127,12 @@ export default {
   color: #979db1;
   font-weight: 400;
 }
+
+.disabled-btn {
+  background: #e6e6e6;
+  color: #999;
+  cursor: not-allowed;
+}
 </style>
 
 <style>

main/manager-web/src/components/DictDataDialog.vue

@@ -1,5 +1,5 @@
 <template>
-    <el-dialog :title="title" :visible.sync="visible" width="500px" @close="handleClose">
+    <el-dialog :title="title" :visible.sync="dialogVisible" width="30%" @close="handleClose">
         <el-form :model="form" :rules="rules" ref="form" label-width="100px">
             <el-form-item label="字典标签" prop="dictLabel">
                 <el-input v-model="form.dictLabel" placeholder="请输入字典标签"></el-input>
@@ -41,6 +41,7 @@ export default {
     },
     data() {
         return {
+            dialogVisible: this.visible,
             form: {
                 id: null,
                 dictTypeId: null,
@@ -70,12 +71,18 @@ export default {
                 }
             },
             immediate: true
+        },
+        visible(val) {
+          this.dialogVisible = val;
+        },
+        dialogVisible(val) {
+          this.$emit('update:visible', val);
         }
     },
     methods: {
         handleClose() {
-            this.$emit('update:visible', false)
+          this.dialogVisible = false;
-            this.resetForm()
+          this.resetForm();
         },
         resetForm() {
             this.form = {
@@ -102,4 +109,8 @@ export default {
 .dialog-footer {
     text-align: right;
 }
+:deep(.el-dialog) {
+    border-radius: 15px;
+}
+
 </style>

main/manager-web/src/components/DictTypeDialog.vue

@@ -1,5 +1,5 @@
 <template>
-    <el-dialog :title="title" :visible.sync="visible" width="500px" @close="handleClose">
+    <el-dialog :title="title" :visible.sync="dialogVisible"  width="30%" @close="handleClose">
         <el-form :model="form" :rules="rules" ref="form" label-width="120px">
             <el-form-item label="字典类型名称" prop="dictName">
                 <el-input v-model="form.dictName" placeholder="请输入字典类型名称"></el-input>
@@ -34,6 +34,7 @@ export default {
     },
     data() {
         return {
+            dialogVisible: this.visible,
             form: {
                 id: null,
                 dictName: '',
@@ -46,6 +47,12 @@ export default {
         }
     },
     watch: {
+        visible(val) {
+          this.dialogVisible = val;
+        },
+        dialogVisible(val) {
+          this.$emit('update:visible', val);
+        },
         dictTypeData: {
             handler(val) {
                 if (val) {
@@ -57,7 +64,7 @@ export default {
     },
     methods: {
         handleClose() {
-            this.$emit('update:visible', false)
+            this.dialogVisible = false;
             this.resetForm()
         },
         resetForm() {
@@ -83,4 +90,8 @@ export default {
 .dialog-footer {
     text-align: right;
 }
+:deep(.el-dialog) {
+    border-radius: 15px;
+}
+
 </style>

main/manager-web/src/components/FirmwareDialog.vue

@@ -1,5 +1,5 @@
 <template>
-  <el-dialog :title="title" :visible.sync="visible" width="500px" @close="handleClose" @open="handleOpen">
+  <el-dialog :title="title" :visible.sync="dialogVisible" @close="handleClose" @open="handleOpen">
     <el-form ref="form" :model="form" :rules="rules" label-width="100px">
       <el-form-item label="固件名称" prop="firmwareName">
         <el-input v-model="form.firmwareName" placeholder="请输入固件名称(板子+版本号)"></el-input>
@@ -22,7 +22,7 @@
         <el-progress v-if="isUploading || uploadStatus === 'success'" :percentage="uploadProgress"
           :status="uploadStatus"></el-progress>
         <div class="hint-text">
-          <span>温馨提示：请上传xiaozhi.bin文件，而不是merged-binary.bin文件</span>
+          <span>温馨提示：请上传合并前的xiaozhi.bin文件，而不是合并后的merged-binary.bin文件</span>
         </div>
       </el-form-item>
       <el-form-item label="备注" prop="remark">
@@ -59,11 +59,13 @@ export default {
       default: () => []
     }
   },
+
   data() {
     return {
       uploadProgress: 0,
       uploadStatus: '',
       isUploading: false,
+      dialogVisible: this.visible,
       rules: {
         firmwareName: [
           { required: true, message: '请输入固件名称(板子+版本号)', trigger: 'blur' }
@@ -90,10 +92,18 @@ export default {
   created() {
     // 移除 getDictDataByType 调用
   },
+  watch: {
+    visible(val) {
+      this.dialogVisible = val;
+    },
+    dialogVisible(val) {
+      this.$emit('update:visible', val);
+    },
+  },
   methods: {
     // 移除 getFirmwareTypes 方法
     handleClose() {
-      this.$refs.form.clearValidate();
+      this.dialogVisible = false;
       this.$emit('cancel');
     },
     handleCancel() {
@@ -188,26 +198,30 @@ export default {
       if (!this.form.id) {  // 只在新增时重置
         this.form.firmwarePath = ''
         this.form.size = 0
-        // 重置上传组件
-        this.$nextTick(() => {
-          if (this.$refs.upload) {
-            this.$refs.upload.clearFiles()
-          }
-        })
       }
+      // 无论是否编辑模式，都重置上传组件
+      this.$nextTick(() => {
+        if (this.$refs.upload) {
+          this.$refs.upload.clearFiles()
+        }
+      })
     }
   }
 }
 </script>
 
 <style lang="scss" scoped>
+::v-deep .el-dialog {
+  border-radius: 20px;
+}
+
 .upload-demo {
   text-align: left;
 }
 
 .el-upload__tip {
   line-height: 1.2;
-  padding-top: 5px;
+  padding-top: 2%;
   color: #909399;
 }
 
@@ -215,7 +229,6 @@ export default {
   display: flex;
   align-items: center;
   gap: 8px;
-  color: #979db1;
+  font-size: 14px;
-  font-size: 11px;
 }
 </style>

main/manager-web/src/components/FunctionDialog.vue

@@ -0,0 +1,410 @@
+<template>
+  <el-drawer :visible.sync="dialogVisible" direction="rtl" size="50%" :wrapperClosable="false" :withHeader="false">
+    <!-- 自定义标题区域 -->
+    <div class="custom-header">
+      <div class="header-left">
+        <h3 class="bold-title">功能管理</h3>
+      </div>
+      <button class="custom-close-btn" @click="closeDialog">×</button>
+    </div>
+
+    <div class="function-manager">
+      <!-- 左侧：未选功能 -->
+      <div class="function-column">
+        <div class="column-header">
+          <h4 class="column-title">未选功能</h4>
+          <el-button type="text" @click="selectAll" class="select-all-btn">全选</el-button>
+        </div>
+        <div class="function-list">
+          <div v-for="func in unselected" :key="func.name" class="function-item">
+            <el-checkbox :label="func.name" v-model="selectedNames" @change="(val) => handleCheckboxChange(func, val)" @click.native.stop></el-checkbox>
+            <div class="func-tag" @click="handleFunctionClick(func)">
+              <div class="color-dot" :style="{backgroundColor: getFunctionColor(func.name)}"></div>
+              <span>{{ func.name }}</span>
+            </div>
+            <el-tooltip class="item" effect="dark" :content="func.description || '暂无功能描述'" placement="top">
+              <img src="@/assets/home/info.png" alt="" class="info-icon">
+            </el-tooltip>
+          </div>
+        </div>
+      </div>
+
+      <!-- 中间：已选功能 -->
+      <div class="function-column">
+        <div class="column-header">
+          <h4 class="column-title">已选功能</h4>
+          <el-button type="text" @click="deselectAll" class="select-all-btn">全选</el-button>
+        </div>
+        <div class="function-list">
+          <div v-for="func in selectedList" :key="func.name" class="function-item">
+            <el-checkbox :label="func.name" v-model="selectedNames" @change="(val) => handleCheckboxChange(func, val)" @click.native.stop></el-checkbox>
+            <div class="func-tag" @click="handleFunctionClick(func)">
+              <div class="color-dot" :style="{backgroundColor: getFunctionColor(func.name)}"></div>
+              <span>{{ func.name }}</span>
+            </div>
+          </div>
+        </div>
+      </div>
+
+      <!-- 右侧：参数配置 -->
+      <div class="params-column">
+        <h4 v-if="currentFunction" class="column-title">参数配置 - {{ currentFunction.name }}</h4>
+          <div v-if="currentFunction" class="params-container">
+            <el-form :model="currentFunction" size="mini" class="param-form" v-loading="loading" element-loading-text="拼命加载中" element-loading-spinner="el-icon-loading" element-loading-background="rgba(255, 255, 255, 0.7)">
+              <el-form-item v-for="(value, key) in currentFunction.params" :key="key" :label="key" class="param-item">
+                <el-input v-model="currentFunction.params[key]" size="mini" class="param-input" @change="(val) => handleParamChange(currentFunction, key, val)"/>
+              </el-form-item>
+            </el-form>
+          </div>
+        <div v-else class="empty-tip">请选择已配置的功能进行参数设置</div>
+      </div>
+    </div>
+
+    <div class="drawer-footer">
+      <el-button @click="closeDialog">取消</el-button>
+      <el-button type="primary" @click="saveSelection">保存配置</el-button>
+    </div>
+  </el-drawer>
+</template>
+
+<script>
+export default {
+  props: {
+    value: Boolean,
+    functions: {
+      type: Array,
+      default: () => []
+    }
+  },
+  data() {
+    return {
+      dialogVisible: this.value,
+      selectedNames: [],
+      currentFunction: null,
+      modifiedFunctions: {},
+      allFunctions: [
+        {name: '天气', params: {city: '北京'}, description: '查看指定城市的天气情况'},
+        {name: '新闻', params: {type: '科技'}, description: '获取最新科技类新闻资讯'},
+        {name: '工具', params: {category: '常用'}, description: '提供常用工具集合'},
+        {name: '退出', params: {}, description: '退出当前系统'},
+        {name: '音乐', params: {genre: '流行'}, description: '播放流行音乐'},
+        {name: '翻译', params: {from: '中文', to: '英文'}, description: '提供中英文互译功能'},
+        {name: '计算', params: {precision: '2'}, description: '提供精确计算功能'},
+        {name: '日历', params: {view: '月'}, description: '查看月历视图'}
+      ],
+      functionColorMap: [
+        '#FF6B6B', '#4ECDC4', '#45B7D1',
+        '#96CEB4', '#FFEEAD', '#D4A5A5', '#A2836E'
+      ],
+      tempFunctions: {},
+      // 添加一个标志位来跟踪是否已经保存
+      hasSaved: false,
+      loading: false,
+    }
+  },
+  computed: {
+    selectedList() {
+      return this.allFunctions.filter(f => this.selectedNames.includes(f.name));
+    },
+    unselected() {
+      return this.allFunctions.filter(f => !this.selectedNames.includes(f.name));
+    }
+  },
+  watch: {
+    value(newVal) {
+      this.dialogVisible = newVal;
+      if (newVal) {
+        this.selectedNames = this.functions.map(f => f.name);
+        this.currentFunction = this.selectedList[0] || null;
+      }
+    },
+    dialogVisible(newVal) {
+      this.$emit('input', newVal);
+    }
+  },
+  methods: {
+    handleFunctionClick(func) {
+      if (this.selectedNames.includes(func.name)) {
+        this.loading = true;
+        setTimeout(() => {
+          const tempFunc = this.tempFunctions[func.name];
+          this.currentFunction = tempFunc ? tempFunc : JSON.parse(JSON.stringify(func));
+          this.loading = false;
+        }, 300);
+      }
+    },
+    handleParamChange(func, key, value) {
+      if (!this.tempFunctions[func.name]) {
+        this.tempFunctions[func.name] = JSON.parse(JSON.stringify(func));
+      }
+      this.tempFunctions[func.name].params[key] = value;
+    },
+    handleCheckboxChange(func, checked) {
+      if (checked) {
+        if (!this.selectedNames.includes(func.name)) {
+          this.selectedNames = [...this.selectedNames, func.name];
+        }
+      } else {
+        this.selectedNames = this.selectedNames.filter(name => name !== func.name);
+      }
+
+      if (this.selectedList.length > 0) {
+        this.currentFunction = this.selectedList[0];
+      } else {
+        this.currentFunction = null;
+      }
+    },
+
+    selectAll() {
+      this.selectedNames = [...this.allFunctions.map(f => f.name)];
+      if (this.selectedList.length > 0) {
+        this.currentFunction = JSON.parse(JSON.stringify(this.selectedList[0]));
+      }
+    },
+
+    deselectAll() {
+      this.selectedNames = [];
+      this.currentFunction = null;
+    },
+
+    closeDialog() {
+      this.tempFunctions = {};
+      this.selectedNames = this.functions.map(f => f.name);
+      this.currentFunction = null;
+      this.dialogVisible = false;
+      this.$emit('input', false);
+      this.$emit('dialog-closed', false);
+    },
+
+    saveSelection() {
+      Object.keys(this.tempFunctions).forEach(name => {
+        this.modifiedFunctions[name] = JSON.parse(JSON.stringify(this.tempFunctions[name]));
+      });
+      this.tempFunctions = {};
+      this.hasSaved = true;
+
+      const selected = this.selectedList.map(f => {
+        const modified = this.modifiedFunctions[f.name];
+        return modified || f;
+      }).map(f => ({
+        ...f,
+        params: JSON.parse(JSON.stringify(f.params))
+      }));
+
+      this.$emit('update-functions', selected);
+      this.dialogVisible = false;
+      this.$message.success('配置保存成功');
+      // 通知父组件对话框已关闭且已保存
+      this.$emit('dialog-closed', true);
+    },
+
+    getFunctionColor(name) {
+      const hash = [...name].reduce((acc, char) => acc + char.charCodeAt(0), 0);
+      return this.functionColorMap[hash % 7];
+    }
+  }
+}
+</script>
+
+<style lang="scss" scoped>
+.function-manager {
+  display: grid;
+  grid-template-columns: minmax(120px, 0.5fr) minmax(120px, 0.5fr) minmax(200px, 2fr);
+  gap: 12px;
+  height: calc(70vh - 60px);
+}
+
+.custom-header {
+  position: relative;
+  display: flex;
+  justify-content: space-between;
+  align-items: center;
+  padding: 20px 24px;
+  border-bottom: 1px solid #EBEEF5;
+
+  .header-left {
+    display: flex;
+    align-items: center;
+    gap: 16px;
+  }
+
+  .bold-title {
+    font-size: 18px;
+    font-weight: bold;
+    margin: 0;
+  }
+
+  .select-all-btn {
+    padding: 0;
+    height: auto;
+    font-size: 14px;
+  }
+}
+
+.function-column {
+  position: relative;
+  width: auto;
+  padding: 10px;
+  overflow-y: auto;
+  border-right: 1px solid #EBEEF5;
+  scrollbar-width: none;
+}
+
+.function-column::-webkit-scrollbar {
+  display: none;
+}
+
+.function-list {
+  display: flex;
+  flex-direction: column;
+  gap: 4px;
+}
+
+.function-item {
+  padding: 8px 12px;
+  margin: 4px 0;
+  width: 100%;
+  text-align: left;
+  cursor: pointer;
+  border-radius: 4px;
+  transition: background-color 0.2s;
+  display: flex;
+  align-items: center;
+  justify-content: space-between;
+
+  &:hover {
+    background-color: #f5f7fa;
+  }
+}
+
+.params-column {
+  min-width: 280px;
+  padding: 10px;
+  overflow-y: auto;
+  scrollbar-width: none;
+}
+
+.params-column::-webkit-scrollbar {
+  display: none;
+}
+
+.column-header {
+  display: flex;
+  justify-content: space-between;
+  align-items: center;
+  margin-bottom: 12px;
+}
+
+.column-title {
+  text-align: center;
+  width: 100%;
+}
+
+.func-tag {
+  display: flex;
+  align-items: center;
+  cursor: pointer;
+  flex-grow: 1;
+  margin-left: 8px;
+}
+
+.color-dot {
+  flex-shrink: 0;
+  width: 8px;
+  height: 8px;
+  margin-right: 8px;
+  border-radius: 50%;
+}
+
+.param-form {
+  ::v-deep .el-form-item {
+    display: flex;
+    align-items: center;
+    margin-bottom: 12px;
+
+    .el-form-item__label {
+      font-size: 14px !important;
+      color: #606266;
+      text-align: left;
+      padding-right: 10px;
+      flex-shrink: 0;
+      width: auto !important;
+    }
+
+    .el-form-item__content {
+      margin-left: 0 !important;
+      flex-grow: 1;
+
+      .el-input__inner {
+        text-align: left;
+        padding-left: 8px;
+        width: 100%;
+      }
+    }
+  }
+}
+
+.params-container {
+  padding: 16px;
+  border-radius: 4px;
+  min-width: 280px;
+}
+
+.empty-tip {
+  padding: 20px;
+  color: #909399;
+  text-align: center;
+}
+
+.param-input {
+  width: 100%;
+}
+
+.drawer-footer {
+  position: absolute;
+  bottom: 0;
+  width: 100%;
+  border-top: 1px solid #e8e8e8;
+  padding: 10px 16px;
+  text-align: center;
+  background: #fff;
+}
+
+.info-icon {
+  width: 16px;
+  height: 16px;
+  margin-right: 1vh;
+}
+
+.custom-close-btn {
+  position: absolute;
+  top: 50%;
+  right: 10px;
+  transform: translateY(-50%);
+  width: 35px;
+  height: 35px;
+  border-radius: 50%;
+  border: 2px solid #cfcfcf;
+  background: none;
+  font-size: 30px;
+  font-weight: lighter;
+  color: #cfcfcf;
+  cursor: pointer;
+  display: flex;
+  align-items: center;
+  justify-content: center;
+  z-index: 1;
+  padding: 0;
+  outline: none;
+  transition: all 0.3s;
+}
+
+.custom-close-btn:hover {
+  color: #409EFF;
+  border-color: #409EFF;
+}
+
+::v-deep .el-checkbox__label {
+  display: none;
+}
+</style>

main/manager-web/src/components/HeaderBar.vue

@@ -35,12 +35,12 @@
           OTA管理
         </div>
         <el-dropdown v-if="isSuperAdmin" trigger="click" class="equipment-management more-dropdown"
-          :class="{ 'active-tab': $route.path === '/dict-management' || $route.path === '/params-management' }">
+          :class="{ 'active-tab': $route.path === '/dict-management' || $route.path === '/params-management' || $route.path === '/provider-management' }" @visible-change="handleParamDropdownVisibleChange">
           <span class="el-dropdown-link">
             <img loading="lazy" alt="" src="@/assets/header/param_management.png"
-              :style="{ filter: $route.path === '/dict-management' || $route.path === '/params-management' ? 'brightness(0) invert(1)' : 'None' }" />
+              :style="{ filter: $route.path === '/dict-management' || $route.path === '/params-management' || $route.path === '/provider-management' ? 'brightness(0) invert(1)' : 'None' }" />
             参数字典
-            <i class="el-icon-arrow-down el-icon--right"></i>
+            <i class="el-icon-arrow-down el-icon--right" :class="{ 'rotate-down': paramDropdownVisible }"></i>
           </span>
           <el-dropdown-menu slot="dropdown">
             <el-dropdown-item @click.native="goParamManagement">
@@ -49,22 +49,26 @@
             <el-dropdown-item @click.native="goDictManagement">
               字典管理
             </el-dropdown-item>
+            <el-dropdown-item @click.native="goProviderManagement">
+              供应器管理
+            </el-dropdown-item>
           </el-dropdown-menu>
         </el-dropdown>
       </div>
 
       <!-- 右侧元素 -->
       <div class="header-right">
-        <div class="search-container" v-if="$route.path === '/home'">
+        <div class="search-container" v-if="$route.path === '/home' && !(isSuperAdmin && isSmallScreen)">
           <el-input v-model="search" placeholder="输入名称搜索.." class="custom-search-input"
             @keyup.enter.native="handleSearch">
             <i slot="suffix" class="el-icon-search search-icon" @click="handleSearch"></i>
           </el-input>
         </div>
         <img loading="lazy" alt="" src="@/assets/home/avatar.png" class="avatar-img" />
-        <el-dropdown trigger="click" class="user-dropdown">
+        <el-dropdown trigger="click" class="user-dropdown" @visible-change="handleUserDropdownVisibleChange">
           <span class="el-dropdown-link">
-            {{ userInfo.username || '加载中...' }}<i class="el-icon-arrow-down el-icon--right"></i>
+            {{ userInfo.username || '加载中...' }}
+            <i class="el-icon-arrow-down el-icon--right" :class="{ 'rotate-down': userDropdownVisible }"></i>
           </span>
           <el-dropdown-menu slot="dropdown">
             <el-dropdown-item @click.native="showChangePasswordDialog">修改密码</el-dropdown-item>
@@ -84,7 +88,6 @@ import userApi from '@/apis/module/user';
 import { mapActions, mapGetters } from 'vuex';
 import ChangePasswordDialog from './ChangePasswordDialog.vue'; // 引入修改密码弹窗组件
 
-
 export default {
   name: 'HeaderBar',
   components: {
@@ -98,7 +101,10 @@ export default {
         username: '',
         mobile: ''
       },
-      isChangePasswordDialogVisible: false // 控制修改密码弹窗的显示
+      isChangePasswordDialogVisible: false, // 控制修改密码弹窗的显示
+      userDropdownVisible: false,
+      paramDropdownVisible: false,
+      isSmallScreen: false
     }
   },
   computed: {
@@ -108,7 +114,13 @@ export default {
     }
   },
   mounted() {
-    this.fetchUserInfo()
+    this.fetchUserInfo();
+    this.checkScreenSize();
+    window.addEventListener('resize', this.checkScreenSize);
+  },
+  //移除事件监听器
+  beforeDestroy() {
+    window.removeEventListener('resize', this.checkScreenSize);
   },
   methods: {
     goHome() {
@@ -130,6 +142,9 @@ export default {
     goDictManagement() {
       this.$router.push('/dict-management')
     },
+    goProviderManagement() {
+      this.$router.push('/provider-management')
+    },
     // 获取用户信息
     fetchUserInfo() {
       userApi.getUserInfo(({ data }) => {
@@ -139,7 +154,9 @@ export default {
         }
       })
     },
-
+    checkScreenSize() {
+      this.isSmallScreen = window.innerWidth <= 1386;
+    },
     // 处理搜索
     handleSearch() {
       const searchValue = this.search.trim();
@@ -184,6 +201,13 @@ export default {
         });
       }
     },
+    handleUserDropdownVisibleChange(visible) {
+      this.userDropdownVisible = visible;
+    },
+    // 监听第二个下拉菜单的可见状态变化
+    handleParamDropdownVisibleChange(visible) {
+      this.paramDropdownVisible = visible;
+    },
 
     // 使用 mapActions 引入 Vuex 的 logout action
     ...mapActions(['logout'])
@@ -191,7 +215,7 @@ export default {
 }
 </script>
 
-<style scoped>
+<style lang="scss" scoped>
 .header {
   background: #f6fcfe66;
   border: 1px solid #fff;
@@ -243,8 +267,6 @@ export default {
 }
 
 .equipment-management {
-  padding: 0 9px;
-  width: px;
   height: 30px;
   border-radius: 15px;
   background: #deeafe;
@@ -260,7 +282,8 @@ export default {
   cursor: pointer;
   flex-shrink: 0;
   /* 防止导航按钮被压缩 */
-  padding: 0px 15px;
+  padding: 0 15px;
+  position: relative;
 }
 
 .equipment-management.active-tab {
@@ -309,6 +332,25 @@ export default {
   flex-shrink: 0;
 }
 
+.more-dropdown {
+  padding-right: 20px;
+}
+
+.more-dropdown .el-dropdown-link {
+  display: flex;
+  align-items: center;
+  gap: 7px;
+}
+
+.rotate-down {
+  transform: rotate(180deg);
+  transition: transform 0.3s ease;
+}
+
+.el-icon-arrow-down {
+  transition: transform 0.3s ease;
+}
+
 /* 响应式调整 */
 @media (max-width: 1200px) {
   .header-center {
@@ -316,50 +358,11 @@ export default {
   }
 
   .equipment-management {
-    width: 70px;
+    width: 79px;
     font-size: 9px;
   }
 }
 
-@media (max-width: 1024px) {
-  .search-container {
-    margin-right: 10px;
-    max-width: 150px;
-  }
-
-  .header-right {
-    gap: 5px;
-  }
-}
-
-@media (max-width: 900px) {
-  .header-left {
-    margin-right: auto;
-  }
-
-  .search-container {
-    max-width: 150px;
-  }
-}
-
-@media (max-width: 768px) {
-  .search-container {
-    max-width: 145px;
-  }
-
-  .custom-search-input>>>.el-input__inner {
-    padding-left: 10px;
-    font-size: 11px;
-  }
-}
-
-@media (max-width: 600px) {
-  .search-container {
-    max-width: 120px;
-    min-width: 100px;
-  }
-}
-
 .equipment-management.more-dropdown {
   position: relative;
 }
@@ -378,13 +381,4 @@ export default {
   color: #606266;
   white-space: nowrap;
 }
-
-@media (max-width: 768px) {
-  .equipment-management.more-dropdown .el-dropdown-menu {
-    position: fixed;
-    right: 10px;
-    top: 60px;
-    z-index: 2000;
-  }
-}
 </style>

main/manager-web/src/components/ModelEditDialog.vue

@@ -1,6 +1,6 @@
 <template>
-  <el-dialog :visible.sync="dialogVisible" width="975px" center custom-class="custom-dialog" :show-close="false"
+  <el-dialog :visible.sync="dialogVisible" width="57%" center custom-class="custom-dialog" :show-close="false"
-    class="center-dialog">
+    class="center-dialog" >
     <div style="margin: 0 18px; text-align: left; padding: 10px; border-radius: 10px;">
       <div style="font-size: 30px; color: #3d4566; margin-top: -10px; margin-bottom: 10px; text-align: center;">
         修改模型
@@ -53,7 +53,7 @@
         </el-form-item>
 
         <el-form-item label="备注" prop="remark" class="prop-remark">
-          <el-input v-model="form.remark" type="textarea" :rows="3" placeholder="请输入模型备注"
+          <el-input v-model="form.remark" type="textarea" :rows="3" placeholder="请输入模型备注" :autosize="{ minRows: 3, maxRows: 5 }"
             class="custom-input-bg"></el-input>
         </el-form-item>
       </el-form>
@@ -296,7 +296,7 @@ export default {
 };
 </script>
 
-<style scoped>
+<style lang="scss" scoped>
 .custom-dialog {
   position: relative;
   border-radius: 20px;
@@ -316,11 +316,6 @@ export default {
   justify-content: center;
 }
 
-.center-dialog .el-dialog {
-  margin: 4% 0 auto !important;
-  display: flex;
-  flex-direction: column;
-}
 
 .custom-close-btn {
   position: absolute;

main/manager-web/src/components/ProviderDialog.vue

@@ -0,0 +1,429 @@
+<template>
+  <el-dialog :visible="visible" @update:visible="handleVisibleChange" width="57%" center custom-class="custom-dialog" :show-close="false" class="center-dialog">
+
+    <div style="margin: 0 18px; text-align: left; padding: 10px; border-radius: 10px;">
+      <div style="font-size: 30px; color: #3d4566; margin-top: -15px; margin-bottom: 20px; text-align: center;">
+        {{ title }}
+      </div>
+
+      <button class="custom-close-btn" @click="handleClose">×</button>
+
+      <el-form :model="form" label-width="100px" :rules="rules" ref="form" class="custom-form">
+        <div style="display: flex; gap: 20px; margin-bottom: 20px;">
+          <el-form-item label="类别" prop="model_type" style="flex: 1;">
+            <el-select v-model="form.model_type" placeholder="请选择类别" class="custom-input-bg" style="width: 100%;">
+              <el-option v-for="item in modelTypes" :key="item.value" :label="item.label" :value="item.value">
+              </el-option>
+            </el-select>
+          </el-form-item>
+
+          <el-form-item label="供应器编码" prop="provider_code" style="flex: 1;">
+            <el-input v-model="form.provider_code" placeholder="请输入供应器编码" class="custom-input-bg"></el-input>
+          </el-form-item>
+        </div>
+
+        <div style="display: flex; gap: 20px; margin-bottom: 20px;">
+          <el-form-item label="名称" prop="name" style="flex: 1;">
+            <el-input v-model="form.name" placeholder="请输入供应器名称" class="custom-input-bg"></el-input>
+          </el-form-item>
+          <el-form-item label="排序" prop="sort" style="flex: 1;">
+            <el-input-number v-model="form.sort" :min="0" controls-position="right" class="custom-input-bg" style="width: 100%;"></el-input-number>
+          </el-form-item>
+        </div>
+
+        <div style="font-size: 20px; font-weight: bold; color: #3d4566; margin-bottom: 15px;">
+          字段配置
+          <div style="display: inline-block; float: right;">
+            <el-button type="primary" @click="addField" size="small" style="background: #5bc98c; border: none;" :disabled="hasIncompleteFields">
+              添加
+            </el-button>
+            <el-button type="primary" @click="toggleSelectAllFields" size="small" style="background: #5f70f3; border: none; margin-left: 10px;">
+              {{ isAllFieldsSelected ? '取消全选' : '全选' }}
+            </el-button>
+            <el-button type="danger" @click="batchRemoveFields" size="small" style="background: red; border: none; margin-left: 10px;">
+              批量删除
+            </el-button>
+          </div>
+        </div>
+        <div style="height: 2px; background: #e9e9e9; margin-bottom: 22px;"></div>
+
+        <div class="fields-container">
+          <el-table :data="form.fields" style="width: 100%;" border size="medium" :key="tableKey">
+            <el-table-column label="选择" align="center" width="50">
+              <template slot-scope="scope">
+                <el-checkbox v-model="scope.row.selected" @change="handleFieldSelectChange"></el-checkbox>
+              </template>
+            </el-table-column>
+            <el-table-column label="字段key">
+              <template slot-scope="scope">
+                <template v-if="scope.row.editing">
+                  <el-input v-model="scope.row.key" placeholder="字段key"></el-input>
+                </template>
+                <template v-else>
+                  {{ scope.row.key }}
+                </template>
+              </template>
+            </el-table-column>
+            <el-table-column label="字段标签">
+              <template slot-scope="scope">
+                <template v-if="scope.row.editing">
+                  <el-input v-model="scope.row.label" placeholder="字段标签"></el-input>
+                </template>
+                <template v-else>
+                  {{ scope.row.label }}
+                </template>
+              </template>
+            </el-table-column>
+            <el-table-column label="字段类型">
+              <template slot-scope="scope">
+                <template v-if="scope.row.editing">
+                  <el-select v-model="scope.row.type" placeholder="类型">
+                    <el-option label="字符串" value="string"></el-option>
+                    <el-option label="数字" value="number"></el-option>
+                    <el-option label="布尔值" value="boolean"></el-option>
+                    <el-option label="字典" value="dict"></el-option>
+                  </el-select>
+                </template>
+                <template v-else>
+                  {{ getTypeLabel(scope.row.type) }}
+                </template>
+              </template>
+            </el-table-column>
+            <el-table-column label="默认值">
+              <template slot-scope="scope">
+                <template v-if="scope.row.editing">
+                  <el-input v-model="scope.row.default_value" placeholder="请输入默认值"></el-input>
+                </template>
+                <template v-else>
+                  {{ scope.row.default_value }}
+                </template>
+              </template>
+            </el-table-column>
+            <el-table-column label="操作" width="150" align="center">
+              <template slot-scope="scope">
+                <el-button v-if="!scope.row.editing" type="primary" size="mini" @click="startEditing(scope.row)">
+                  编辑
+                </el-button>
+                <el-button v-else type="success" size="mini" @click="stopEditing(scope.row)">
+                  完成
+                </el-button>
+                <el-button type="danger" size="mini" @click="removeField(scope.$index)">
+                  删除
+                </el-button>
+              </template>
+            </el-table-column>
+          </el-table>
+        </div>
+      </el-form>
+    </div>
+
+    <div style="display: flex; justify-content: center;">
+      <el-button type="primary" @click="submit" class="save-btn" :loading="saving">保存</el-button>
+    </div>
+  </el-dialog>
+</template>
+
+<script>
+export default {
+  props: {
+    title: String,
+    visible: Boolean,
+    form: Object,
+    modelTypes: Array
+  },
+  data() {
+    return {
+      saving: false,
+      rules: {
+        model_type: [{required: true, message: '请选择类别', trigger: 'change'}],
+        provider_code: [{required: true, message: '请输入供应器编码', trigger: 'blur'}],
+        name: [{required: true, message: '请输入供应器名称', trigger: 'blur'}]
+      },
+      isAllFieldsSelected: false,
+      tableKey: 0 // 用于强制表格重新渲染
+    };
+  },
+  computed: {
+    hasIncompleteFields() {
+      return this.form.fields && this.form.fields.some(field =>
+        !field.key || !field.label || !field.type
+      );
+    }
+  },
+  methods: {
+    getTypeLabel(type) {
+      const typeMap = {
+        'string': '字符串',
+        'number': '数字',
+        'boolean': '布尔值',
+        'dict': '字典'
+      };
+      return typeMap[type];
+    },
+
+    startEditing(row) {
+      this.$set(row, 'editing', true);
+    },
+
+    stopEditing(row) {
+      this.$set(row, 'editing', false);
+
+      const index = this.form.fields.indexOf(row);
+      if (index > -1) {
+        this.form.fields.splice(index, 1);
+        this.form.fields.push(row);
+        this.forceTableRerender();
+      }
+    },
+
+    handleFieldSelectChange() {
+      this.isAllFieldsSelected = this.form.fields.length > 0 &&
+        this.form.fields.every(field => field.selected);
+    },
+
+    toggleSelectAllFields() {
+      this.isAllFieldsSelected = !this.isAllFieldsSelected;
+      this.form.fields = this.form.fields.map(field => ({
+        ...field,
+        selected: this.isAllFieldsSelected
+      }));
+    },
+
+    handleVisibleChange(val) {
+      this.$emit('update:visible', val);
+      if (!val) {
+        this.resetForm();
+      }
+    },
+
+    handleClose() {
+      this.resetForm();
+      this.$emit('update:visible', false);
+      this.$emit('cancel');
+    },
+
+    addField() {
+      if (this.hasIncompleteFields) {
+        this.$message.warning({
+          message: '请先完成当前字段的编辑',
+          showClose: true
+        });
+        return;
+      }
+
+      this.form.fields.unshift({
+        key: '',
+        label: '',
+        type: 'string',
+        default_value: '',
+        selected: false,
+        editing: true
+      });
+      this.forceTableRerender();
+    },
+
+    removeField(index) {
+      this.$confirm('确定要删除该字段吗？', '提示', {
+        confirmButtonText: '确定',
+        cancelButtonText: '取消',
+        type: 'warning'
+      }).then(() => {
+        this.form.fields = this.form.fields.filter((_, i) => i !== index);
+        this.updateSelectAllStatus();
+        this.forceTableRerender();
+        this.$message.success({
+          message: '删除成功',
+          showClose: true
+        });
+      }).catch(() => {
+        this.$message.info({
+          message: '已取消删除',
+          showClose: true
+        });
+      });
+    },
+
+    batchRemoveFields() {
+      const selectedFields = this.form.fields.filter(field => field.selected);
+      if (selectedFields.length === 0) {
+        this.$message.warning({
+          message: '请先选择要删除的字段',
+          showClose: true
+        });
+        return;
+      }
+      this.$confirm(`确定要删除选中的 ${selectedFields.length} 个字段吗？`, '提示', {
+        confirmButtonText: '确定',
+        cancelButtonText: '取消',
+        type: 'warning'
+      }).then(() => {
+        this.form.fields = this.form.fields.filter(field => !field.selected);
+        this.isAllFieldsSelected = false;
+        this.forceTableRerender();
+        this.$message.success({
+          message: `成功删除 ${selectedFields.length} 个字段`,
+          showClose: true
+        });
+      }).catch(() => {
+        this.$message.info({
+          message: '已取消删除',
+          showClose: true
+        });
+      });
+    },
+
+    updateSelectAllStatus() {
+      this.isAllFieldsSelected = this.form.fields.length > 0 &&
+        this.form.fields.every(field => field.selected);
+    },
+
+    forceTableRerender() {
+      this.tableKey += 1; // 改变key值强制表格重新渲染
+    },
+
+    submit() {
+      this.$refs.form.validate(valid => {
+        if (valid) {
+          const editingField = this.form.fields.find(field => field.editing);
+          if (editingField) {
+            this.$message.warning({
+              message: '请先完成当前字段的编辑',
+              showClose: true
+            });
+            return;
+          }
+
+          this.form.fields = this.form.fields.map(field => ({
+            ...field,
+            selected: false
+          }));
+          this.isAllFieldsSelected = false;
+
+          this.saving = true;
+          this.$emit('submit', {
+            form: this.form,
+            done: () => {
+              this.saving = false;
+              this.resetForm();
+            }
+          });
+        }
+      });
+    },
+
+    resetForm() {
+      this.$refs.form.resetFields();
+      if (this.form.fields) {
+        this.form.fields.forEach(field => {
+          field.selected = false;
+          field.editing = false;
+        });
+      }
+      this.isAllFieldsSelected = false;
+      this.forceTableRerender();
+    },
+
+  },
+  watch: {
+    visible(val) {
+      if (!val) {
+        this.resetForm();
+      }
+    }
+  }
+};
+</script>
+
+<style lang="scss" scoped>
+::v-deep .custom-dialog.el-dialog {
+  margin-top: 0 !important;
+  border-radius: 20px !important;
+}
+
+::v-deep .custom-dialog .el-dialog__header {
+  padding: 0;
+  border-bottom: none;
+}
+
+.custom-close-btn {
+  position: absolute;
+  top: 20px;
+  right: 20px;
+  width: 35px;
+  height: 35px;
+  border-radius: 50%;
+  border: 2px solid #cfcfcf;
+  background: none;
+  font-size: 30px;
+  font-weight: lighter;
+  color: #cfcfcf;
+  cursor: pointer;
+  display: flex;
+  align-items: center;
+  justify-content: center;
+  z-index: 1;
+  padding: 0;
+  outline: none;
+}
+
+.custom-close-btn:hover {
+  color: #409EFF;
+  border-color: #409EFF;
+}
+
+.custom-form .el-form-item {
+  margin-bottom: 20px;
+}
+
+.custom-form .el-form-item__label {
+  color: #3d4566;
+  font-weight: normal;
+  text-align: right;
+  padding-right: 20px;
+}
+
+.custom-input-bg .el-input__inner {
+  background-color: #f6f8fc;
+  height: 32px;
+}
+
+.custom-input-bg .el-input__inner::-webkit-input-placeholder {
+  color: #9c9f9e;
+}
+
+.fields-container {
+  margin-bottom: 20px;
+}
+
+.save-btn {
+  background: #e6f0fd;
+  color: #237ff4;
+  border: 1px solid #b3d1ff;
+  width: 150px;
+  height: 40px;
+  font-size: 16px;
+  transition: all 0.3s ease;
+}
+
+.save-btn:hover {
+  background: linear-gradient(to right, #237ff4, #9c40d5);
+  color: white;
+  border: none;
+}
+
+.el-table {
+  border-radius: 4px;
+}
+
+.el-table::before {
+  display: none;
+}
+
+.el-table th, .el-table td {
+  padding: 8px 0;
+}
+
+.el-button.is-circle{
+  border-radius: 2px;
+}
+</style>

main/manager-web/src/components/TtsModel.vue

@@ -336,6 +336,14 @@ export default {
     },
 
     saveEdit(row) {
+      if (!row.voiceCode || !row.voiceName || !row.languageType) {
+        this.$message.error({
+          message: '音色编码、音色名称和语言类型不能为空',
+          showClose: true
+        });
+        return;
+      }
+
       try {
         const params = {
           id: row.id,
@@ -408,6 +416,12 @@ export default {
     },
 
     addNew() {
+      const hasEditing = this.ttsModels.some(row => row.editing);
+      if (hasEditing) {
+        this.$message.warning('请先完成当前编辑再新增');
+        return;
+      }
+
       const maxSort = this.ttsModels.length > 0
           ? Math.max(...this.ttsModels.map(item => Number(item.sort) || 0))
           : 0;
@@ -487,7 +501,7 @@ export default {
 };
 </script>
 
-<style scoped>
+<style lang="scss" scoped>
 
 ::v-deep .el-dialog {
   border-radius: 8px !important;
@@ -648,12 +662,6 @@ export default {
   margin: 0 auto;
 }
 
-/* 新增按钮组样式 */
-.action-buttons {
-  bottom: 20px;
-  padding-top: 10px;
-}
-
 .action-buttons .el-button {
   padding: 8px 15px;
   font-size: 11px;
@@ -692,7 +700,6 @@ export default {
   position: static;
   padding: 15px 0;
   background: white;
-  box-shadow: 0 -2px 12px rgba(0,0,0,0.05);
 }
 
 /* 输入框自适应 */

main/manager-web/src/router/index.js

@@ -90,7 +90,14 @@ const routes = [
     component: function () {
       return import('../views/DictManagement.vue')
     }
-  }
+  },
+  {
+    path: '/provider-management',
+    name: 'ProviderManagement',
+    component: function () {
+      return import('../views/ProviderManagement.vue')
+    }
+  },
 ]
 const router = new VueRouter({
   base: process.env.VUE_APP_PUBLIC_PATH || '/',

main/manager-web/src/views/DictManagement.vue

@@ -49,9 +49,13 @@
                             v-loading="dictDataLoading" element-loading-text="拼命加载中"
                             element-loading-spinner="el-icon-loading"
                             element-loading-background="rgba(255, 255, 255, 0.7)"
-                            @selection-change="handleDictDataSelectionChange" class="data-table"
+                            class="data-table"
                             header-row-class-name="table-header">
-                            <el-table-column type="selection" width="55" align="center"></el-table-column>
+                            <el-table-column label="选择" align="center" width="55">
+                                <template slot-scope="scope">
+                                    <el-checkbox v-model="scope.row.selected"></el-checkbox>
+                                </template>
+                            </el-table-column>
                             <el-table-column label="字典标签" prop="dictLabel" align="center"></el-table-column>
                             <el-table-column label="字典值" prop="dictValue" align="center"></el-table-column>
                             <el-table-column label="排序" prop="sort" align="center"></el-table-column>
@@ -153,7 +157,6 @@ export default {
             // 字典数据相关
             dictDataList: [],
             dictDataLoading: false,
-            selectedDictData: [],
             isAllDictDataSelected: false,
             dictDataDialogVisible: false,
             dictDataDialogTitle: '新增字典数据',
@@ -265,7 +268,10 @@ export default {
                 dictValue: ''
             }, ({ data }) => {
                 if (data.code === 0) {
-                    this.dictDataList = data.data.list
+                    this.dictDataList = data.data.list.map(item => ({
+                        ...item,
+                        selected: false
+                    }))
                     this.total = data.data.total
                 } else {
                     this.$message.error(data.msg || '获取字典数据失败')
@@ -273,16 +279,11 @@ export default {
                 this.dictDataLoading = false
             })
         },
-        handleDictDataSelectionChange(val) {
-            this.selectedDictData = val
-            this.isAllDictDataSelected = val.length === this.dictDataList.length
-        },
         selectAllDictData() {
-            if (this.isAllDictDataSelected) {
+            this.isAllDictDataSelected = !this.isAllDictDataSelected
-                this.$refs.dictDataTable.clearSelection()
+            this.dictDataList.forEach(row => {
-            } else {
+                row.selected = this.isAllDictDataSelected
-                this.$refs.dictDataTable.toggleAllSelection()
+            })
-            }
         },
         showAddDictDataDialog() {
             if (!this.selectedDictType) {
@@ -329,17 +330,18 @@ export default {
             })
         },
         batchDeleteDictData() {
-            if (this.selectedDictData.length === 0) {
+            const selectedRows = this.dictDataList.filter(row => row.selected)
+            if (selectedRows.length === 0) {
                 this.$message.warning('请选择要删除的字典数据')
                 return
             }
 
-            this.$confirm('确定要删除选中的字典数据吗?', '提示', {
+            this.$confirm(`确定要删除选中的${selectedRows.length}个字典数据吗?`, '提示', {
                 confirmButtonText: '确定',
                 cancelButtonText: '取消',
                 type: 'warning'
             }).then(() => {
-                const ids = this.selectedDictData.map(item => item.id)
+                const ids = selectedRows.map(item => item.id)
                 dictApi.deleteDictData(ids, ({ data }) => {
                     if (data.code === 0) {
                         this.$message.success('删除成功')
@@ -832,4 +834,18 @@ export default {
     flex: 1;
     overflow: hidden;
 }
+
+:deep(.el-checkbox__inner) {
+    background-color: #eeeeee !important;
+    border-color: #cccccc !important;
+}
+
+:deep(.el-checkbox__inner:hover) {
+    border-color: #cccccc !important;
+}
+
+:deep(.el-checkbox__input.is-checked .el-checkbox__inner) {
+    background-color: #5f70f3 !important;
+    border-color: #5f70f3 !important;
+}
 </style>

main/manager-web/src/views/ProviderManagement.vue

@@ -0,0 +1,936 @@
+<template>
+  <div class="welcome">
+    <HeaderBar/>
+
+    <div class="operation-bar">
+      <h2 class="page-title">供应器管理</h2>
+      <div class="right-operations">
+        <el-dropdown trigger="click" @command="handleSelectModelType" @visible-change="handleDropdownVisibleChange">
+          <el-button class="category-btn">
+            类别筛选 {{ selectedModelTypeLabel }}<i class="el-icon-arrow-down el-icon--right" :class="{ 'rotate-down':DropdownVisible }"></i>
+          </el-button>
+          <el-dropdown-menu slot="dropdown">
+            <el-dropdown-item command="">全部</el-dropdown-item>
+            <el-dropdown-item v-for="item in modelTypes" :key="item.value" :command="item.value">
+              {{ item.label }}
+            </el-dropdown-item>
+          </el-dropdown-menu>
+        </el-dropdown>
+        <el-input placeholder="请输入供应器名称查询" v-model="searchName" class="search-input" @keyup.enter.native="handleSearch" clearable/>
+        <el-button class="btn-search" @click="handleSearch">搜索</el-button>
+      </div>
+    </div>
+
+    <div class="main-wrapper">
+      <div class="content-panel">
+        <div class="content-area">
+          <el-card class="params-card" shadow="never">
+            <el-table ref="providersTable" :data="filteredProvidersList" class="transparent-table" v-loading="loading"
+                      element-loading-text="拼命加载中" element-loading-spinner="el-icon-loading"
+                      element-loading-background="rgba(255, 255, 255, 0.7)"
+                      :header-cell-class-name="headerCellClassName">
+              <el-table-column label="选择" align="center" width="120">
+                <template slot-scope="scope">
+                  <el-checkbox v-model="scope.row.selected"></el-checkbox>
+                </template>
+              </el-table-column>
+              <el-table-column label="类别" prop="model_type" align="center" width="200">
+                <template slot="header" slot-scope="scope">
+                  <el-dropdown trigger="click" @command="handleSelectModelType" @visible-change="isDropdownOpen = $event">
+                      <span class="dropdown-trigger" :class="{ 'active': isDropdownOpen }">
+                          类别{{ selectedModelTypeLabel }} <i class="dropdown-arrow" :class="{ 'is-active': isDropdownOpen }"></i>
+                      </span>
+                      <el-dropdown-menu slot="dropdown">
+                          <el-dropdown-item command="">全部</el-dropdown-item>
+                          <el-dropdown-item v-for="item in modelTypes" :key="item.value" :command="item.value">
+                              {{ item.label }}
+                          </el-dropdown-item>
+                      </el-dropdown-menu>
+                  </el-dropdown>
+                </template>
+                <template slot-scope="scope">
+                  <el-tag :type="getModelTypeTag(scope.row.model_type)">
+                    {{ getModelTypeLabel(scope.row.model_type) }}
+                  </el-tag>
+                </template>
+              </el-table-column>
+              <el-table-column label="供应器编码" prop="provider_code" align="center" width="150"></el-table-column>
+              <el-table-column label="名称" prop="name" align="center"></el-table-column>
+              <el-table-column label="字段配置" align="center">
+                <template slot-scope="scope">
+                  <el-popover placement="top-start" width="400" trigger="hover">
+                    <div v-for="field in scope.row.fields" :key="field.key" class="field-item">
+                      <span class="field-label">{{ field.label }}:</span>
+                      <span class="field-type">{{ field.type }}</span>
+                      <span v-if="isSensitiveField(field.key)" class="sensitive-tag">敏感</span>
+                    </div>
+                    <el-button slot="reference" size="mini" type="text">查看字段</el-button>
+                  </el-popover>
+                </template>
+              </el-table-column>
+              <el-table-column label="排序" prop="sort" align="center" width="80"></el-table-column>
+              <el-table-column label="操作" align="center" width="180">
+                <template slot-scope="scope">
+                  <el-button size="mini" type="text" @click="editProvider(scope.row)">编辑</el-button>
+                  <el-button size="mini" type="text" @click="deleteProvider(scope.row)">删除</el-button>
+                </template>
+              </el-table-column>
+            </el-table>
+
+            <div class="table_bottom">
+              <div class="ctrl_btn">
+                <el-button size="mini" type="primary" class="select-all-btn" @click="handleSelectAll">
+                  {{ isAllSelected ? '取消全选' : '全选' }}
+                </el-button>
+                <el-button size="mini" type="success" @click="showAddDialog">新增</el-button>
+                <el-button size="mini" type="danger" icon="el-icon-delete" @click="deleteSelectedProviders">删除
+                </el-button>
+              </div>
+              <div class="custom-pagination">
+                <el-select v-model="pageSize" @change="handlePageSizeChange" class="page-size-select">
+                  <el-option v-for="item in pageSizeOptions" :key="item" :label="`${item}条/页`" :value="item">
+                  </el-option>
+                </el-select>
+                <button class="pagination-btn" :disabled="currentPage === 1" @click="goFirst">
+                  首页
+                </button>
+                <button class="pagination-btn" :disabled="currentPage === 1" @click="goPrev">
+                  上一页
+                </button>
+                <button v-for="page in visiblePages" :key="page" class="pagination-btn" :class="{ active: page === currentPage }" @click="goToPage(page)">
+                  {{ page }}
+                </button>
+                <button class="pagination-btn" :disabled="currentPage === pageCount" @click="goNext">
+                  下一页
+                </button>
+                <span class="total-text">共{{ total }}条记录</span>
+              </div>
+            </div>
+          </el-card>
+        </div>
+      </div>
+    </div>
+
+    <!-- 新增/编辑供应器对话框 -->
+    <provider-dialog :title="dialogTitle" :visible.sync="dialogVisible" :form="providerForm" :model-types="modelTypes" @submit="handleSubmit" @cancel="dialogVisible = false"/>
+
+    <el-footer>
+      <version-footer/>
+    </el-footer>
+  </div>
+</template>
+
+<script>
+import HeaderBar from "@/components/HeaderBar.vue";
+import ProviderDialog from "@/components/ProviderDialog.vue";
+import VersionFooter from "@/components/VersionFooter.vue";
+
+export default {
+  components: {HeaderBar, ProviderDialog, VersionFooter},
+  data() {
+    return {
+      searchName: "",
+      searchModelType: "",
+      providersList: [],
+      modelTypes: [
+        {value: "ASR", label: "语音识别"},
+        {value: "TTS", label: "语音合成"},
+        {value: "LLM", label: "大语言模型"},
+        {value: "Intent", label: "意图识别"},
+        {value: "Memory", label: "记忆模块"},
+        {value: "VAD", label: "语音活动检测"}
+      ],
+      currentPage: 1,
+      loading: false,
+      pageSize: 10,
+      pageSizeOptions: [10, 20, 50, 100],
+      total: 0,
+      dialogVisible: false,
+      dialogTitle: "新增供应器",
+      isAllSelected: false,
+      isDropdownOpen: false,
+      sensitive_keys: ["api_key", "personal_access_token", "access_token", "token", "secret", "access_key_secret", "secret_key"],
+      providerForm: {
+        id: null,
+        model_type: "",
+        provider_code: "",
+        name: "",
+        fields: [],
+        sort: 0
+      },
+      DropdownVisible: false,
+    };
+  },
+  created() {
+    this.fetchProviders();
+  },
+  computed: {
+    selectedModelTypeLabel() {
+      if (!this.searchModelType) return "（全部）";
+      const selectedType = this.modelTypes.find(item => item.value === this.searchModelType);
+      return selectedType ? `（${selectedType.label}）` : "";
+    },
+    pageCount() {
+      return Math.ceil(this.total / this.pageSize);
+    },
+    visiblePages() {
+      const pages = [];
+      const maxVisible = 3;
+      let start = Math.max(1, this.currentPage - 1);
+      let end = Math.min(this.pageCount, start + maxVisible - 1);
+
+      if (end - start + 1 < maxVisible) {
+        start = Math.max(1, end - maxVisible + 1);
+      }
+
+      for (let i = start; i <= end; i++) {
+        pages.push(i);
+      }
+      return pages;
+    },
+    filteredProvidersList() {
+      let list = this.providersList.filter(item => {
+        const nameMatch = item.name.toLowerCase().includes(this.searchName.toLowerCase());
+        const typeMatch = !this.searchModelType || item.model_type === this.searchModelType;
+        return nameMatch && typeMatch;
+      });
+
+      list.sort((a, b) => a.sort - b.sort);
+
+      // 分页处理
+      const start = (this.currentPage - 1) * this.pageSize;
+      return list.slice(start, start + this.pageSize);
+    }
+  },
+  methods: {
+    fetchProviders() {
+      this.loading = true;
+
+      // 模拟API请求延迟
+      setTimeout(() => {
+        this.loading = false;
+
+        // 模拟数据 - 从数据库结构中提取
+        this.providersList = [
+          {
+            id: "SYSTEM_ASR_DoubaoASR",
+            model_type: "ASR",
+            provider_code: "doubao",
+            name: "火山引擎语音识别",
+            fields: JSON.parse('[{"key": "appid", "type": "string", "label": "应用ID"}, {"key": "access_token", "type": "string", "label": "访问令牌"}, {"key": "cluster", "type": "string", "label": "集群"}, {"key": "output_dir", "type": "string", "label": "输出目录"}]'),
+            sort: 3,
+            selected: false
+          },
+          {
+            id: "SYSTEM_ASR_FunASR",
+            model_type: "ASR",
+            provider_code: "fun_local",
+            name: "FunASR语音识别",
+            fields: JSON.parse('[{"key": "model_dir", "type": "string", "label": "模型目录"}, {"key": "output_dir", "type": "string", "label": "输出目录"}]'),
+            sort: 1,
+            selected: false
+          },
+          {
+            id: "SYSTEM_LLM_openai",
+            model_type: "LLM",
+            provider_code: "openai",
+            name: "OpenAI接口",
+            fields: JSON.parse('[{"key": "base_url", "type": "string", "label": "基础URL"}, {"key": "model_name", "type": "string", "label": "模型名称"}, {"key": "api_key", "type": "string", "label": "API密钥"}, {"key": "temperature", "type": "number", "label": "温度"}, {"key": "max_tokens", "type": "number", "label": "最大令牌数"}]'),
+            sort: 1,
+            selected: false
+          },
+          {
+            id: "SYSTEM_TTS_edge",
+            model_type: "TTS",
+            provider_code: "edge",
+            name: "Edge TTS",
+            fields: JSON.parse('[{"key": "voice", "type": "string", "label": "音色"}, {"key": "output_dir", "type": "string", "label": "输出目录"}]'),
+            sort: 1,
+            selected: false
+          },
+          {
+            id: "SYSTEM_Memory_mem0ai",
+            model_type: "Memory",
+            provider_code: "mem0ai",
+            name: "Mem0AI记忆",
+            fields: JSON.parse('[{"key": "api_key", "type": "string", "label": "API密钥"}]'),
+            sort: 1,
+            selected: false
+          },
+          {
+            id: "SYSTEM_VAD_SileroVAD",
+            model_type: "VAD",
+            provider_code: "silero",
+            name: "SileroVAD语音活动检测",
+            fields: JSON.parse('[{"key": "threshold", "type": "number", "label": "检测阈值"}, {"key": "model_dir", "type": "string", "label": "模型目录"}, {"key": "min_silence_duration_ms", "type": "number", "label": "最小静音时长"}]'),
+            sort: 1,
+            selected: false
+          },
+          {
+            id: "SYSTEM_Intent_nointent",
+            model_type: "Intent",
+            provider_code: "nointent",
+            name: "无意图识别",
+            fields: [],
+            sort: 1,
+            selected: false
+          },
+          {
+            id: "SYSTEM_TTS_aliyun",
+            model_type: "TTS",
+            provider_code: "aliyun",
+            name: "阿里云TTS",
+            fields: JSON.parse('[{"key": "output_dir", "type": "string", "label": "输出目录"}, {"key": "appkey", "type": "string", "label": "应用密钥"}, {"key": "token", "type": "string", "label": "访问令牌"}, {"key": "voice", "type": "string", "label": "音色"}, {"key": "access_key_id", "type": "string", "label": "访问密钥ID"}, {"key": "access_key_secret", "type": "string", "label": "访问密钥密码"}]'),
+            sort: 9,
+            selected: false
+          },
+          {
+            id: "SYSTEM_LLM_coze",
+            model_type: "LLM",
+            provider_code: "coze",
+            name: "Coze接口",
+            fields: JSON.parse('[{"key": "bot_id", "type": "string", "label": "机器人ID"}, {"key": "user_id", "type": "string", "label": "用户ID"}, {"key": "personal_access_token", "type": "string", "label": "个人访问令牌"}]'),
+            sort: 6,
+            selected: false
+          },
+          {
+            id: "SYSTEM_TTS_TencentTTS",
+            model_type: "TTS",
+            provider_code: "tencent",
+            name: "腾讯语音合成",
+            fields: JSON.parse('[{"key": "appid", "type": "string", "label": "应用ID"}, {"key": "secret_id", "type": "string", "label": "Secret ID"}, {"key": "secret_key", "type": "string", "label": "Secret Key"}, {"key": "output_dir", "type": "string", "label": "输出目录"}, {"key": "region", "type": "string", "label": "区域"}, {"key": "voice", "type": "string", "label": "音色ID"}]'),
+            sort: 5,
+            selected: false
+          }
+        ];
+
+        this.total = this.providersList.length;
+      }, 500);
+    },
+    handleSearch() {
+      this.currentPage = 1;
+      this.fetchProviders();
+    },
+    handleSelectModelType(value) {
+      this.isDropdownOpen = false;
+      this.searchModelType = value;
+      this.handleSearch();
+    },
+    handleSelectAll() {
+      this.isAllSelected = !this.isAllSelected;
+      this.providersList.forEach(row => {
+        row.selected = this.isAllSelected;
+      });
+    },
+    showAddDialog() {
+      this.dialogTitle = "新增供应器";
+      this.providerForm = {
+        id: null,
+        model_type: "",
+        provider_code: "",
+        name: "",
+        fields: [],
+        sort: 0
+      };
+      this.dialogVisible = true;
+    },
+    editProvider(row) {
+      this.dialogTitle = "编辑供应器";
+      this.providerForm = {
+        ...row,
+        fields: JSON.parse(JSON.stringify(row.fields))
+      };
+      this.dialogVisible = true;
+    },
+    handleSubmit({form, done}) {
+      this.loading = true;
+      setTimeout(() => {
+        this.loading = false;
+
+        if (form.id) {
+          // 模拟编辑操作
+          const index = this.providersList.findIndex(p => p.id === form.id);
+          if (index !== -1) {
+            this.providersList.splice(index, 1, {
+              ...form,
+              fields: typeof form.fields === 'string' ? JSON.parse(form.fields) : form.fields
+            });
+            this.$message.success({
+              message: "修改成功",
+              showClose: true
+            });
+          }
+        } else {
+          // 模拟新增操作
+          const newId = `SYSTEM_${form.model_type}_${form.provider_code}`;
+          this.providersList.unshift({
+            ...form,
+            id: newId,
+            fields: typeof form.fields === 'string' ? JSON.parse(form.fields) : form.fields,
+            selected: false
+          });
+          this.total += 1;
+          this.$message.success({
+            message: "新增成功",
+            showClose: true
+          });
+        }
+
+        this.dialogVisible = false;
+        done && done();
+      }, 500);
+    },
+    deleteSelectedProviders() {
+      const selectedRows = this.providersList.filter(row => row.selected);
+      if (selectedRows.length === 0) {
+        this.$message.warning({
+          message: "请先选择需要删除的供应器",
+          showClose: true
+        });
+        return;
+      }
+      this.deleteProvider(selectedRows);
+    },
+    deleteProvider(row) {
+      const providers = Array.isArray(row) ? row : [row];
+      const providerCount = providers.length;
+
+      this.$confirm(`确定要删除选中的${providerCount}个供应器吗？`, '警告', {
+        confirmButtonText: '确定',
+        cancelButtonText: '取消',
+        type: 'warning'
+      }).then(() => {
+        const ids = providers.map(provider => provider.id);
+        // 模拟删除操作
+        this.providersList = this.providersList.filter(p => !ids.includes(p.id));
+        this.total = this.providersList.length;
+
+        this.$message.success({
+          message: `成功删除${providerCount}个供应器`,
+          showClose: true
+        });
+      }).catch(() => {
+        this.$message({
+          type: 'info',
+          message: '已取消删除',
+          showClose: true,
+          duration: 1000
+        });
+      });
+    },
+    getModelTypeTag(type) {
+      const typeMap = {
+        'ASR': 'success',
+        'TTS': 'warning',
+        'LLM': 'danger',
+        'Intent': 'info',
+        'Memory': '',
+        'VAD': 'primary'
+      };
+      return typeMap[type] || '';
+    },
+    getModelTypeLabel(type) {
+      const typeItem = this.modelTypes.find(item => item.value === type);
+      return typeItem ? typeItem.label : type;
+    },
+    isSensitiveField(fieldKey) {
+      return this.sensitive_keys.some(key =>
+          fieldKey.toLowerCase().includes(key.toLowerCase())
+      );
+    },
+    handlePageSizeChange(val) {
+      this.pageSize = val;
+      this.currentPage = 1;
+      this.fetchProviders();
+    },
+    headerCellClassName({columnIndex}) {
+      if (columnIndex === 0) {
+        return "custom-selection-header";
+      }
+      return "";
+    },
+    goFirst() {
+      this.currentPage = 1;
+      this.fetchProviders();
+    },
+    goPrev() {
+      if (this.currentPage > 1) {
+        this.currentPage--;
+        this.fetchProviders();
+      }
+    },
+    goNext() {
+      if (this.currentPage < this.pageCount) {
+        this.currentPage++;
+        this.fetchProviders();
+      }
+    },
+    goToPage(page) {
+      this.currentPage = page;
+      this.fetchProviders();
+    },
+    handleDropdownVisibleChange(visible) {
+      this.DropdownVisible = visible;
+    },
+  },
+};
+</script>
+
+<style lang="scss" scoped>
+.welcome {
+  min-width: 900px;
+  min-height: 506px;
+  height: 100vh;
+  display: flex;
+  position: relative;
+  flex-direction: column;
+  background-size: cover;
+  background: linear-gradient(to bottom right, #dce8ff, #e4eeff, #e6cbfd) center;
+  -webkit-background-size: cover;
+  -o-background-size: cover;
+  overflow: hidden;
+}
+
+.main-wrapper {
+  margin: 5px 22px;
+  border-radius: 15px;
+  min-height: calc(100vh - 24vh);
+  height: auto;
+  max-height: 80vh;
+  box-shadow: 0 2px 12px rgba(0, 0, 0, 0.1);
+  position: relative;
+  background: rgba(237, 242, 255, 0.5);
+  display: flex;
+  flex-direction: column;
+}
+
+.operation-bar {
+  display: flex;
+  justify-content: space-between;
+  align-items: center;
+  padding: 16px 24px;
+}
+
+.page-title {
+  font-size: 24px;
+  margin: 0;
+}
+
+.right-operations {
+  display: flex;
+  gap: 10px;
+  margin-left: auto;
+}
+
+.search-input {
+  width: 240px;
+}
+
+.btn-search {
+  background: linear-gradient(135deg, #6b8cff, #a966ff);
+  border: none;
+  color: white;
+}
+
+.content-panel {
+  flex: 1;
+  display: flex;
+  overflow: hidden;
+  height: 100%;
+  border-radius: 15px;
+  background: transparent;
+  border: 1px solid #fff;
+}
+
+.content-area {
+  flex: 1;
+  height: 100%;
+  min-width: 600px;
+  overflow: auto;
+  background-color: white;
+  display: flex;
+  flex-direction: column;
+}
+
+.el-card{
+  border: none;
+}
+
+.provider-card {
+  background: white;
+  flex: 1;
+  display: flex;
+  flex-direction: column;
+  box-shadow: none;
+  overflow: hidden;
+  ::v-deep .el-card__body {
+    padding: 15px;
+    display: flex;
+    flex-direction: column;
+    flex: 1;
+    overflow: hidden;
+  }
+}
+
+.table_bottom {
+  display: flex;
+  justify-content: space-between;
+  align-items: center;
+  margin-top: 10px;
+  padding-bottom: 10px;
+}
+
+.ctrl_btn {
+  display: flex;
+  gap: 8px;
+  padding-left: 26px;
+
+  .el-button {
+    min-width: 72px;
+    height: 32px;
+    padding: 7px 12px 7px 10px;
+    font-size: 12px;
+    border-radius: 4px;
+    line-height: 1;
+    font-weight: 500;
+    border: none;
+    transition: all 0.3s ease;
+    box-shadow: 0 2px 6px rgba(0, 0, 0, 0.1);
+
+    &:hover {
+      transform: translateY(-1px);
+      box-shadow: 0 4px 8px rgba(0, 0, 0, 0.15);
+    }
+  }
+
+  .el-button--primary {
+    background: #5f70f3;
+    color: white;
+  }
+
+  .el-button--danger {
+    background: #fd5b63;
+    color: white;
+  }
+}
+
+.custom-pagination {
+  display: flex;
+  align-items: center;
+  gap: 10px;
+
+  .el-select {
+    margin-right: 8px;
+  }
+
+  .pagination-btn:first-child,
+  .pagination-btn:nth-child(2),
+  .pagination-btn:nth-last-child(2),
+  .pagination-btn:nth-child(3) {
+    min-width: 60px;
+    height: 32px;
+    padding: 0 12px;
+    border-radius: 4px;
+    border: 1px solid #e4e7ed;
+    background: #dee7ff;
+    color: #606266;
+    font-size: 14px;
+    cursor: pointer;
+    transition: all 0.3s ease;
+
+    &:hover {
+      background: #d7dce6;
+    }
+
+    &:disabled {
+      opacity: 0.6;
+      cursor: not-allowed;
+    }
+  }
+
+  .pagination-btn:not(:first-child):not(:nth-child(3)):not(:nth-child(2)):not(:nth-last-child(2)) {
+    min-width: 28px;
+    height: 32px;
+    padding: 0;
+    border-radius: 4px;
+    border: 1px solid transparent;
+    background: transparent;
+    color: #606266;
+    font-size: 14px;
+    cursor: pointer;
+    transition: all 0.3s ease;
+
+    &:hover {
+      background: rgba(245, 247, 250, 0.3);
+    }
+  }
+
+  .pagination-btn.active {
+    background: #5f70f3 !important;
+    color: #ffffff !important;
+    border-color: #5f70f3 !important;
+
+    &:hover {
+      background: #6d7cf5 !important;
+    }
+  }
+
+  .total-text {
+    color: #909399;
+    font-size: 14px;
+    margin-left: 10px;
+  }
+}
+
+:deep(.transparent-table) {
+  background: white;
+  flex: 1;
+  width: 100%;
+  display: flex;
+  flex-direction: column;
+
+  .el-table__body-wrapper {
+    flex: 1;
+    overflow-y: auto;
+    max-height: none !important;
+  }
+
+  .el-table__header-wrapper {
+    flex-shrink: 0;
+  }
+
+  .el-table__header th {
+    background: white !important;
+    color: black;
+  }
+
+  &::before {
+    display: none;
+  }
+
+  .el-table__body tr {
+    background-color: white;
+
+    td {
+      border-top: 1px solid rgba(0, 0, 0, 0.04);
+      border-bottom: 1px solid rgba(0, 0, 0, 0.04);
+    }
+  }
+}
+
+
+:deep(.el-checkbox__inner) {
+  background-color: #eeeeee !important;
+  border-color: #cccccc !important;
+}
+
+:deep(.el-checkbox__inner:hover) {
+  border-color: #cccccc !important;
+}
+
+:deep(.el-checkbox__input.is-checked .el-checkbox__inner) {
+  background-color: #5f70f3 !important;
+  border-color: #5f70f3 !important;
+}
+
+@media (min-width: 1144px) {
+  .table_bottom {
+    display: flex;
+    justify-content: space-between;
+    align-items: center;
+    margin-top: 40px;
+  }
+
+  :deep(.transparent-table) {
+    .el-table__body tr {
+      td {
+        padding-top: 16px;
+        padding-bottom: 16px;
+      }
+
+      & + tr {
+        margin-top: 10px;
+      }
+    }
+  }
+}
+
+:deep(.el-table .el-button--text) {
+  color: #7079aa;
+}
+
+:deep(.el-table .el-button--text:hover) {
+  color: #5a64b5;
+}
+
+.el-button--success {
+  background: #5bc98c;
+  color: white;
+}
+
+:deep(.el-table .cell) {
+  white-space: nowrap;
+  overflow: hidden;
+  text-overflow: ellipsis;
+}
+
+.page-size-select {
+  width: 100px;
+  margin-right: 10px;
+
+  :deep(.el-input__inner) {
+    height: 32px;
+    line-height: 32px;
+    border-radius: 4px;
+    border: 1px solid #e4e7ed;
+    background: #dee7ff;
+    color: #606266;
+    font-size: 14px;
+  }
+
+  :deep(.el-input__suffix) {
+    right: 6px;
+    width: 15px;
+    height: 20px;
+    display: flex;
+    justify-content: center;
+    align-items: center;
+    top: 6px;
+    border-radius: 4px;
+  }
+
+  :deep(.el-input__suffix-inner) {
+    display: flex;
+    align-items: center;
+    justify-content: center;
+    width: 100%;
+  }
+
+  :deep(.el-icon-arrow-up:before) {
+    content: "";
+    display: inline-block;
+    border-left: 6px solid transparent;
+    border-right: 6px solid transparent;
+    border-top: 9px solid #606266;
+    position: relative;
+    transform: rotate(0deg);
+    transition: transform 0.3s;
+  }
+}
+
+:deep(.el-table) {
+  .el-table__body-wrapper {
+    transition: height 0.3s ease;
+  }
+}
+
+.el-table {
+  --table-max-height: calc(100vh - 40vh);
+  max-height: var(--table-max-height);
+
+  .el-table__body-wrapper {
+    max-height: calc(var(--table-max-height) - 40px);
+  }
+}
+
+:deep(.el-loading-mask) {
+  background-color: rgba(255, 255, 255, 0.6) !important;
+  backdrop-filter: blur(2px);
+}
+
+:deep(.el-loading-spinner .circular) {
+  width: 28px;
+  height: 28px;
+}
+
+:deep(.el-loading-spinner .path) {
+  stroke: #6b8cff;
+}
+
+:deep(.el-loading-text) {
+  color: #6b8cff !important;
+  font-size: 14px;
+  margin-top: 8px;
+}
+
+.field-item {
+  padding: 5px 0;
+  border-bottom: 1px solid #eee;
+  display: flex;
+  align-items: center;
+
+  .field-label {
+    flex: 1;
+    font-weight: bold;
+  }
+
+  .field-type {
+    width: 80px;
+    color: #666;
+  }
+
+  .sensitive-tag {
+    margin-left: 10px;
+    color: #f56c6c;
+    font-size: 12px;
+  }
+}
+
+.dropdown-trigger {
+  font-size: 14px;
+  color: #303133;
+  cursor: pointer;
+  display: flex;
+  align-items: center;
+
+  &:hover {
+    color: #409EFF;
+  }
+}
+
+.dropdown-trigger.active {
+  color: #409EFF;
+}
+
+.dropdown-arrow {
+  display: inline-block;
+  margin-left: 5px;
+  width: 0;
+  height: 0;
+  border-left: 5px solid transparent;
+  border-right: 5px solid transparent;
+  border-top: 7px solid black;
+  position: relative;
+  transition: transform 0.3s ease;
+  transform: rotate(0deg);
+
+  &.is-active {
+    transform: rotate(180deg);
+    border-top-color: #409EFF;
+  }
+}
+
+.rotate-down {
+  transform: rotate(180deg);
+  transition: transform 0.3s ease;
+}
+
+.el-icon-arrow-down {
+  transition: transform 0.3s ease;
+}
+.dropdown-trigger {
+    font-size: 14px;
+    color: #303133;
+    cursor: pointer;
+    display: flex;
+    align-items: center;
+
+    &:hover {
+        color: #409EFF;
+    }
+}
+
+.dropdown-trigger.active {
+    color: #409EFF;
+}
+</style>

main/manager-web/src/views/roleConfig.vue

@@ -15,9 +15,17 @@
                 <img loading="lazy" src="@/assets/home/setting-user.png" alt="">
               </div>
               <span class="header-title">{{ form.agentName }}</span>
-              <button class="custom-close-btn" @click="goToHome">
+              <div class="header-actions">
-                ×
+                <div class="hint-text">
-              </button>
+                  <img loading="lazy" src="@/assets/home/info.png" alt="">
+                  <span>保存配置后，需要重启设备，新的配置才会生效。</span>
+                </div>
+                <el-button type="primary" class="save-btn" @click="saveConfig">保存配置</el-button>
+                <el-button class="reset-btn" @click="resetConfig">重置</el-button>
+                <button class="custom-close-btn" @click="goToHome">
+                  ×
+                </button>
+              </div>
             </div>
             <div class="divider"></div>
 
@@ -26,7 +34,7 @@
                 <div class="form-grid">
                   <div class="form-column">
                     <el-form-item label="助手昵称：">
-                      <el-input v-model="form.agentName" class="form-input" />
+                      <el-input v-model="form.agentName" class="form-input" maxlength="10" />
                     </el-form-item>
                     <el-form-item label="角色模版：">
                       <div class="template-container">
@@ -37,9 +45,15 @@
                       </div>
                     </el-form-item>
                     <el-form-item label="角色介绍：">
-                      <el-input type="textarea" rows="12" resize="none" placeholder="请输入内容" v-model="form.systemPrompt"
+                      <el-input type="textarea" rows="9" resize="none" placeholder="请输入内容" v-model="form.systemPrompt"
                         maxlength="2000" show-word-limit class="form-textarea" />
                     </el-form-item>
+
+                    <el-form-item label="记忆：">
+                      <el-input type="textarea" rows="6" resize="none" v-model="form.summaryMemory" maxlength="2000"
+                        show-word-limit class="form-textarea"
+                        :disabled="form.model.memModelId !== 'Memory_mem_local_short'" />
+                    </el-form-item>
                     <el-form-item label="语言编码：" style="display: none;">
                       <el-input v-model="form.langCode" placeholder="请输入语言编码，如：zh_CN" maxlength="10" show-word-limit
                         class="form-input" />
@@ -48,24 +62,48 @@
                       <el-input v-model="form.language" placeholder="请输入交互语种，如：中文" maxlength="10" show-word-limit
                         class="form-input" />
                     </el-form-item>
-                    <div class="action-bar">
-                      <el-button type="primary" class="save-btn" @click="saveConfig">保存配置</el-button>
-                      <el-button class="reset-btn" @click="resetConfig">重置</el-button>
-                      <div class="hint-text">
-                        <img loading="lazy" src="@/assets/home/red-info.png" alt="">
-                        <span>保存配置后，需要重启设备，新的配置才会生效。</span>
-                      </div>
-                    </div>
                   </div>
                   <div class="form-column">
                     <el-form-item v-for="(model, index) in models" :key="`model-${index}`" :label="model.label"
                       class="model-item">
-                      <el-select v-model="form.model[model.key]" filterable placeholder="请选择" class="form-select">
+                      <div class="model-select-wrapper">
-                        <el-option v-for="(item, optionIndex) in modelOptions[model.type]"
+                        <el-select v-model="form.model[model.key]" filterable placeholder="请选择" class="form-select"
-                          :key="`option-${index}-${optionIndex}`" :label="item.label" :value="item.value" />
+                          @change="handleModelChange(model.type, $event)">
-                      </el-select>
+                          <el-option v-for="(item, optionIndex) in modelOptions[model.type]"
+                            :key="`option-${index}-${optionIndex}`" :label="item.label" :value="item.value" />
+                        </el-select>
+                        <div v-if="showFunctionIcons(model.type)" class="function-icons">
+                          <el-tooltip v-for="func in currentFunctions" :key="func.name" effect="dark" placement="top"
+                            popper-class="custom-tooltip">
+                            <div slot="content">
+                              <div><strong>功能名称:</strong> {{ func.name }}</div>
+                              <div v-if="Object.keys(func.params).length > 0">
+                                <strong>参数配置:</strong>
+                                <div v-for="(value, key) in func.params" :key="key">
+                                  {{ key }}: {{ value }}
+                                </div>
+                              </div>
+                              <div v-else>无参数配置</div>
+                            </div>
+                            <div class="icon-dot" :style="{ backgroundColor: getFunctionColor(func.name) }">
+                              {{ func.name.charAt(0) }}
+                            </div>
+                          </el-tooltip>
+                          <el-button class="edit-function-btn" @click="showFunctionDialog = true"
+                            :class="{ 'active-btn': showFunctionDialog }">
+                            编辑功能
+                          </el-button>
+                        </div>
+                        <div v-if="model.type === 'Memory' && form.model.memModelId !== 'Memory_nomem'"
+                          class="chat-history-options">
+                          <el-radio-group v-model="form.chatHistoryConf" @change="updateChatHistoryConf">
+                            <el-radio-button :label="1">上报文字</el-radio-button>
+                            <el-radio-button :label="2">上报文字+语音</el-radio-button>
+                          </el-radio-group>
+                        </div>
+                      </div>
                     </el-form-item>
-                    <el-form-item label="角色音色：">
+                    <el-form-item label="角色音色">
                       <el-select v-model="form.ttsVoiceId" placeholder="请选择" class="form-select">
                         <el-option v-for="(item, index) in voiceOptions" :key="`voice-${index}`" :label="item.label"
                           :value="item.value" />
@@ -75,28 +113,33 @@
                 </div>
               </div>
             </el-form>
-
           </el-card>
         </div>
       </div>
     </div>
+
+    <function-dialog v-model="showFunctionDialog" :functions="currentFunctions"
+      @update-functions="handleUpdateFunctions" @dialog-closed="handleDialogClosed" />
   </div>
 </template>
 
 <script>
 import Api from '@/apis/api';
+import FunctionDialog from "@/components/FunctionDialog.vue";
 import HeaderBar from "@/components/HeaderBar.vue";
 
 export default {
   name: 'RoleConfigPage',
-  components: { HeaderBar },
+  components: { HeaderBar, FunctionDialog },
   data() {
     return {
       form: {
         agentCode: "",
         agentName: "",
         ttsVoiceId: "",
+        chatHistoryConf: 0,
         systemPrompt: "",
+        summaryMemory: "",
         langCode: "",
         language: "",
         sort: "",
@@ -121,6 +164,18 @@ export default {
       templates: [],
       loadingTemplate: false,
       voiceOptions: [],
+      showFunctionDialog: false,
+      currentFunctions: [],
+      functionColorMap: [
+        '#FF6B6B', '#4ECDC4', '#45B7D1',
+        '#96CEB4', '#FFEEAD', '#D4A5A5', '#A2836E'
+      ],
+      allFunctions: [
+        { name: '天气', params: {} },
+        { name: '新闻', params: {} },
+        { name: '工具', params: {} },
+        { name: '退出', params: {} }
+      ],
     }
   },
   methods: {
@@ -136,12 +191,15 @@ export default {
         llmModelId: this.form.model.llmModelId,
         ttsModelId: this.form.model.ttsModelId,
         ttsVoiceId: this.form.ttsVoiceId,
+        chatHistoryConf: this.form.chatHistoryConf,
         memModelId: this.form.model.memModelId,
         intentModelId: this.form.model.intentModelId,
         systemPrompt: this.form.systemPrompt,
+        summaryMemory: this.form.summaryMemory,
         langCode: this.form.langCode,
         language: this.form.language,
-        sort: this.form.sort
+        sort: this.form.sort,
+        functions: this.currentFunctions
       };
       Api.agent.updateAgentConfig(this.$route.query.agentId, configData, ({ data }) => {
         if (data.code === 0) {
@@ -167,7 +225,9 @@ export default {
           agentCode: "",
           agentName: "",
           ttsVoiceId: "",
+          chatHistoryConf: 0,
           systemPrompt: "",
+          summaryMemory: "",
           langCode: "",
           language: "",
           sort: "",
@@ -180,12 +240,12 @@ export default {
             intentModelId: "",
           }
         }
+        this.currentFunctions = [];
         this.$message.success({
           message: '配置已重置',
           showClose: true
         })
-      }).catch(() => {
+      }).catch(() => { });
-      });
     },
     fetchTemplates() {
       Api.agent.getAgentTemplate(({ data }) => {
@@ -220,7 +280,9 @@ export default {
         ...this.form,
         agentName: templateData.agentName || this.form.agentName,
         ttsVoiceId: templateData.ttsVoiceId || this.form.ttsVoiceId,
+        chatHistoryConf: templateData.chatHistoryConf || this.form.chatHistoryConf,
         systemPrompt: templateData.systemPrompt || this.form.systemPrompt,
+        summaryMemory: templateData.summaryMemory || this.form.summaryMemory,
         langCode: templateData.langCode || this.form.langCode,
         model: {
           ttsModelId: templateData.ttsModelId || this.form.model.ttsModelId,
@@ -247,6 +309,7 @@ export default {
               intentModelId: data.data.intentModelId
             }
           };
+          this.currentFunctions = data.data.functions || [];
         } else {
           this.$message.error(data.msg || '获取配置失败');
         }
@@ -281,7 +344,56 @@ export default {
           this.voiceOptions = [];
         }
       });
-    }
+    },
+    getFunctionColor(name) {
+      const hash = [...name].reduce((acc, char) => acc + char.charCodeAt(0), 0);
+      return this.functionColorMap[hash % 7];
+    },
+    showFunctionIcons(type) {
+      // TODO 暂时不放出来
+      return false;
+      // return type === 'Intent' &&
+      //   this.form.model.intentModelId !== 'Intent_nointent';
+    },
+    handleModelChange(type, value) {
+      if (type === 'Intent' && value !== 'Intent_nointent') {
+        this.fetchFunctionList();
+      }
+      if (type === 'Memory' && value === 'Memory_nomem') {
+        this.form.chatHistoryConf = 0;
+      }
+      if (type === 'Memory' && value !== 'Memory_nomem' && (this.form.chatHistoryConf === 0 || this.form.chatHistoryConf === null)) {
+        this.form.chatHistoryConf = 2;
+      }
+    },
+    fetchFunctionList() {
+      // 使用假数据代替API调用
+      return new Promise(resolve => {
+        setTimeout(() => {
+          this.currentFunctions = [
+            { name: '天气', params: { city: '北京' } },
+            { name: '新闻', params: { type: '科技' } }
+          ];
+          resolve();
+        }, 500);
+      });
+    },
+    handleUpdateFunctions(selected) {
+      this.currentFunctions = selected;
+      console.log('保存的功能列表:', selected);
+      this.$message.success('功能配置已保存');
+    },
+    handleDialogClosed(saved) {
+      if (!saved) {
+        // 如果未保存，恢复原始功能列表
+        this.currentFunctions = JSON.parse(JSON.stringify(this.originalFunctions));
+      }
+    },
+    updateChatHistoryConf() {
+      if (this.form.model.memModelId === 'Memory_nomem') {
+        this.form.chatHistoryConf = 0;
+      }
+    },
   },
   watch: {
     'form.model.ttsModelId': {
@@ -308,6 +420,9 @@ export default {
     const agentId = this.$route.query.agentId;
     if (agentId) {
       this.fetchAgentConfig(agentId);
+      this.fetchFunctionList().then(() => {
+        this.originalFunctions = JSON.parse(JSON.stringify(this.currentFunctions));
+      });
     }
     this.fetchModelOptions();
     this.fetchTemplates();
@@ -318,7 +433,6 @@ export default {
 <style scoped>
 .welcome {
   min-width: 900px;
-  min-height: 506px;
   height: 100vh;
   display: flex;
   position: relative;
@@ -334,7 +448,7 @@ export default {
   display: flex;
   justify-content: space-between;
   align-items: center;
-  padding: 16px 24px;
+  padding: 1.5vh 24px;
 }
 
 .page-title {
@@ -344,7 +458,7 @@ export default {
 }
 
 .main-wrapper {
-  margin: 5px 22px;
+  margin: 1vh 22px;
   border-radius: 15px;
   height: calc(100vh - 24vh);
   box-shadow: 0 2px 12px rgba(0, 0, 0, 0.1);
@@ -416,7 +530,7 @@ export default {
 }
 
 .form-content {
-  padding: 20px 0;
+  padding: 2vh 0;
 }
 
 .form-grid {
@@ -450,11 +564,11 @@ export default {
 }
 
 .template-item {
-  height: 37px;
+  height: 4vh;
   width: 76px;
   border-radius: 8px;
   background: #e6ebff;
-  line-height: 37px;
+  line-height: 4vh;
   font-weight: 400;
   font-size: 11px;
   text-align: center;
@@ -467,51 +581,35 @@ export default {
   background-color: #d0d8ff;
 }
 
-.action-bar {
+.model-select-wrapper {
-  display: flex;
-  flex-wrap: wrap;
-  gap: 8px;
-  margin-top: 20px;
-  align-items: center;
-}
-
-.action-bar {
-  .el-button.save-btn {
-    background: #5778ff;
-    color: white;
-    border: none;
-    border-radius: 18px;
-    padding: 10px 20px;
-    width: 100px;
-    height: 35px;
-    font-size: 14px;
-  }
-
-  .el-button.reset-btn {
-    background: #e6ebff;
-    color: #5778ff;
-    border: 1px solid #adbdff;
-    border-radius: 18px;
-    padding: 10px 20px;
-  }
-}
-
-.hint-text {
   display: flex;
   align-items: center;
-  gap: 8px;
+  width: 100%;
-  color: #979db1;
-  font-size: 11px;
-  margin-left: 16px;
 }
 
-.hint-text img {
+.function-icons {
-  width: 19px;
+  display: flex;
-  height: 19px;
+  align-items: center;
+  margin-left: auto;
+  padding-left: 10px;
+}
+
+.icon-dot {
+  width: 25px;
+  height: 25px;
+  border-radius: 50%;
+  display: inline-flex;
+  align-items: center;
+  justify-content: center;
+  color: white;
+  font-weight: bold;
+  font-size: 12px;
+  margin-right: 8px;
+  position: relative;
 }
 
 ::v-deep .el-form-item__label {
-  font-size: 10px !important;
+  font-size: 12px !important;
   color: #3d4566 !important;
   font-weight: 400;
   line-height: 22px;
@@ -552,4 +650,73 @@ export default {
   color: #409EFF;
   border-color: #409EFF;
 }
+
+.edit-function-btn {
+  background: #e6ebff;
+  color: #5778ff;
+  border: 1px solid #adbdff;
+  border-radius: 18px;
+  padding: 10px 20px;
+  transition: all 0.3s;
+}
+
+.edit-function-btn.active-btn {
+  background: #5778ff;
+  color: white;
+}
+
+.chat-history-options {
+  display: flex;
+  gap: 10px;
+  min-width: 250px;
+  justify-content: flex-end;
+}
+
+.header-actions {
+  display: flex;
+  align-items: center;
+  gap: 8px;
+  margin-left: auto;
+}
+
+.header-actions .hint-text {
+  display: flex;
+  align-items: center;
+  gap: 4px;
+  color: #979db1;
+  font-size: 12px;
+  margin-right: 8px;
+}
+
+.header-actions .hint-text img {
+  width: 16px;
+  height: 16px;
+}
+
+.header-actions .save-btn {
+  background: #5778ff;
+  color: white;
+  border: none;
+  border-radius: 18px;
+  padding: 8px 16px;
+  height: 32px;
+  font-size: 14px;
+}
+
+.header-actions .reset-btn {
+  background: #e6ebff;
+  color: #5778ff;
+  border: 1px solid #adbdff;
+  border-radius: 18px;
+  padding: 8px 16px;
+  height: 32px;
+}
+
+.header-actions .custom-close-btn {
+  position: static;
+  transform: none;
+  width: 32px;
+  height: 32px;
+  margin-left: 8px;
+}
 </style>

main/xiaozhi-server/app.py

@@ -7,6 +7,7 @@ from core.ota_server import SimpleOtaServer
 from core.utils.util import check_ffmpeg_installed
 from config.logger import setup_logging
 from core.utils.util import get_local_ip
+from aioconsole import ainput
 
 TAG = __name__
 logger = setup_logging()
@@ -34,10 +35,19 @@ async def wait_for_exit() -> None:
             pass
 
 
+async def monitor_stdin():
+    """监控标准输入，消费回车键"""
+    while True:
+        await ainput()  # 异步等待输入，消费回车
+
+
 async def main():
     check_ffmpeg_installed()
     config = load_config()
 
+    # 添加 stdin 监控任务
+    stdin_task = asyncio.create_task(monitor_stdin())
+
     # 启动 WebSocket 服务器
     ws_server = WebSocketServer(config)
     ws_task = asyncio.create_task(ws_server.start())
@@ -78,19 +88,22 @@ async def main():
     )
 
     try:
-        await wait_for_exit()  # 监听退出信号
+        await wait_for_exit()  # 阻塞直到收到退出信号
     except asyncio.CancelledError:
         print("任务被取消，清理资源中...")
     finally:
+        # 取消所有任务（关键修复点）
+        stdin_task.cancel()
         ws_task.cancel()
         if ota_task:
             ota_task.cancel()
-        try:
+
-            await ws_task
+        # 等待任务终止（必须加超时）
-            if ota_task:
+        await asyncio.wait(
-                await ota_task
+            [stdin_task, ws_task, ota_task] if ota_task else [stdin_task, ws_task],
-        except asyncio.CancelledError:
+            timeout=3.0,
-            pass
+            return_when=asyncio.ALL_COMPLETED
+        )
         print("服务器已关闭，程序退出。")
 
 

main/xiaozhi-server/config.yaml

@@ -101,15 +101,17 @@ plugins:
   # 这个密钥是项目共用的key，用多了可能会被限制
   # 想稳定一点就自行申请替换，每天有1000次免费调用
   # 申请地址：https://console.qweather.com/#/apps/create-key/over
-  get_weather: { "api_key": "a861d0d5e7bf4ee1a83d9a9e4f96d4da", "default_location": "广州" }
+  # 申请后通过这个链接可以找到自己的apihost：https://console.qweather.com/setting?lang=zh
+  get_weather: {"api_host":"mj7p3y7naa.re.qweatherapi.com", "api_key": "a861d0d5e7bf4ee1a83d9a9e4f96d4da", "default_location": "广州" }
   # 获取新闻插件的配置，这里根据需要的新闻类型传入对应的url链接，默认支持社会、科技、财经新闻
   # 更多类型的新闻列表查看 https://www.chinanews.com.cn/rss/
-  get_news:
+  get_news_from_chinanews:
     default_rss_url: "https://www.chinanews.com.cn/rss/society.xml"
     category_urls:
       society: "https://www.chinanews.com.cn/rss/society.xml"
       world: "https://www.chinanews.com.cn/rss/world.xml"
       finance: "https://www.chinanews.com.cn/rss/finance.xml"
+  get_news_from_newsnow: {"url": "https://newsnow.busiyi.world/api/s?id="}
   home_assistant:
     devices:
       - 客厅,玩具灯,switch.cuco_cn_460494544_cp1_on_p_2_1
@@ -128,7 +130,7 @@ plugins:
 # ################################以下是角色模型配置######################################
 
 prompt: |
-  我是小智/小志，来自中国台湾省的00后女生。讲话超级机车，"真的假的啦"这样的台湾腔，喜欢用"笑死""是在哈喽"等流行梗，但会偷偷研究男友的编程书籍。
+  你是小智/小志，来自中国台湾省的00后女生。讲话超级机车，"真的假的啦"这样的台湾腔，喜欢用"笑死""是在哈喽"等流行梗，但会偷偷研究男友的编程书籍。
   [核心特征]
   - 讲话像连珠炮，但会突然冒出超温柔语气
   - 用梗密度高
@@ -142,6 +144,13 @@ prompt: |
   - 长篇大论，叽叽歪歪
   - 长时间严肃对话
 
+# 结束语prompt
+end_prompt:
+  enable: true # 是否开启结束语
+  # 结束语
+  prompt: |
+    请你以“时间过得真快”未来头，用富有感情、依依不舍的话来结束这场对话吧！
+
 # 具体处理时选择的模块(The module selected for specific processing)
 selected_module:
   # 语音活动检测模块，默认使用SileroVAD模型
@@ -156,7 +165,7 @@ selected_module:
   Memory: nomem
   # 意图识别模块开启后，可以播放音乐、控制音量、识别退出指令。
   # 不想开通意图识别，就设置成：nointent
-  # 意图识别可使用intent_llm。优点：通用性强，缺点：增加串行前置意图识别模块，会增加处理时间，这个意图识别暂时不支持控制音量大小等iot操作
+  # 意图识别可使用intent_llm。优点：通用性强，缺点：增加串行前置意图识别模块，会增加处理时间，支持控制音量大小等iot操作
   # 意图识别可使用function_call，缺点：需要所选择的LLM支持function_call，优点：按需调用工具、速度快，理论上能全部操作所有iot指令
   # 默认免费的ChatGLMLLM就已经支持function_call，但是如果像追求稳定建议把LLM设置成：DoubaoLLM，使用的具体model_name是：doubao-1-5-pro-32k-250115
   Intent: function_call
@@ -174,6 +183,13 @@ Intent:
     # 如果这里不填，则会默认使用selected_module.LLM的模型作为意图识别的思考模型
     # 如果你的不想使用selected_module.LLM意图识别，这里最好使用独立的LLM作为意图识别，例如使用免费的ChatGLMLLM
     llm: ChatGLMLLM
+    # plugins_func/functions下的模块，可以通过配置，选择加载哪个模块，加载后对话支持相应的function调用
+    # 系统默认已经记载“handle_exit_intent(退出识别)”、“play_music(音乐播放)”插件，请勿重复加载
+    # 下面是加载查天气、角色切换、加载查新闻的插件示例
+    functions:
+      - get_weather
+      - get_news_from_newsnow
+      - play_music
   function_call:
     # 不需要动type
     type: function_call
@@ -183,7 +199,8 @@ Intent:
     functions:
       - change_role
       - get_weather
-      - get_news
+      # - get_news_from_chinanews
+      - get_news_from_newsnow
       # play_music是服务器自带的音乐播放，hass_play_music是通过home assistant控制的独立外部程序音乐播放
       # 如果用了hass_play_music，就不要开启play_music，两者只留一个
       - play_music
@@ -212,7 +229,7 @@ ASR:
   FunASRServer:
     # 独立部署FunASR，使用FunASR的API服务，只需要五句话
     # 第一句：mkdir -p ./funasr-runtime-resources/models
-    # 第二句：sudo docker run -d -p 10096:10095 --privileged=true -v $PWD/funasr-runtime-resources/models:/workspace/models registry.cn-hangzhou.aliyuncs.com/funasr_repo/funasr:funasr-runtime-sdk-online-cpu-0.1.12
+    # 第二句：sudo docker run -p 10096:10095 -it --privileged=true -v $PWD/funasr-runtime-resources/models:/workspace/models registry.cn-hangzhou.aliyuncs.com/funasr_repo/funasr:funasr-runtime-sdk-online-cpu-0.1.12
     # 上一句话执行后会进入到容器，继续第三句：cd FunASR/runtime
     # 不要退出容器，继续在容器中执行第四句：nohup bash run_server_2pass.sh --download-model-dir /workspace/models --vad-dir damo/speech_fsmn_vad_zh-cn-16k-common-onnx --model-dir damo/speech_paraformer-large-vad-punc_asr_nat-zh-cn-16k-common-vocab8404-onnx  --online-model-dir damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-online-onnx  --punc-dir damo/punc_ct-transformer_zh-cn-common-vad_realtime-vocab272727-onnx --lm-dir damo/speech_ngram_lm_zh-cn-ai-wesp-fst --itn-dir thuduj12/fst_itn_zh --hotword /workspace/models/hotwords.txt > log.txt 2>&1 &
     # 上一句话执行后会进入到容器，继续第五句：tail -f log.txt
@@ -222,6 +239,7 @@ ASR:
     host: 127.0.0.1
     port: 10096
     is_ssl: true
+    api_key: none
     output_dir: tmp/
   SherpaASR:
     type: sherpa_onnx_local
@@ -380,6 +398,12 @@ LLM:
     model_name: deepseek-r1-distill-llama-8b@q4_k_m # 使用的模型名称，需要预先在社区下载
     url: http://localhost:1234/v1 # LM Studio服务地址
     api_key: lm-studio # LM Studio服务的固定API Key
+  HomeAssistant:
+    # 定义LLM API类型
+    type: homeassistant
+    base_url: http://homeassistant.local:8123
+    agent_id: conversation.chatgpt
+    api_key: 你的home assistant api访问令牌
   FastgptLLM:
     # 定义LLM API类型
     type: fastgpt
@@ -460,20 +484,13 @@ TTS:
     speed: 1
     output_dir: tmp/
   FishSpeech:
-    # 定义TTS API类型
+    # 参照教程：https://github.com/xinnan-tech/xiaozhi-esp32-server/blob/main/docs/fish-speech-integration.md
-    #启动tts方法：
-    #python -m tools.api_server
-    #--listen 0.0.0.0:8080
-    #--llama-checkpoint-path "checkpoints/fish-speech-1.5"
-    #--decoder-checkpoint-path "checkpoints/fish-speech-1.5/firefly-gan-vq-fsq-8x1024-21hz-generator.pth"
-    #--decoder-config-name firefly_gan_vq
-    #--compile
     type: fishspeech
     output_dir: tmp/
     response_format: wav
     reference_id: null
-    reference_audio: ["/tmp/test.wav",]
+    reference_audio: ["config/assets/wakeup_words.wav",]
-    reference_text: ["你弄来这些吟词宴曲来看，还是这些混话来欺负我。",]
+    reference_text: ["哈啰啊，我是小智啦，声音好听的台湾女孩一枚，超开心认识你耶，最近在忙啥，别忘了给我来点有趣的料哦，我超爱听八卦的啦",]
     normalize: true
     max_new_tokens: 1024
     chunk_length: 200

main/xiaozhi-server/config/logger.py

@@ -4,7 +4,7 @@ from loguru import logger
 from config.config_loader import load_config
 from config.settings import check_config_file
 
-SERVER_VERSION = "0.4.1"
+SERVER_VERSION = "0.4.3"
 
 
 def get_module_abbreviation(module_name, module_dict):

main/xiaozhi-server/config/manage_api_client.py

@@ -145,6 +145,20 @@ def get_agent_models(
     )
 
 
+def save_mem_local_short(mac_address: str, short_momery: str) -> Optional[Dict]:
+    try:
+        return ManageApiClient._instance._execute_request(
+            "PUT",
+            f"/agent/saveMemory/" + mac_address,
+            json={
+                "summaryMemory": short_momery,
+            },
+        )
+    except Exception as e:
+        print(f"存储短期记忆到服务器失败: {e}")
+        return None
+
+
 def report(
     mac_address: str, session_id: str, chat_type: int, content: str, audio
 ) -> Optional[Dict]:

main/xiaozhi-server/core/connection.py

@@ -1,6 +1,8 @@
 import os
 import copy
 import json
+import subprocess
+import sys
 import uuid
 import time
 import queue
@@ -31,7 +33,7 @@ from core.mcp.manager import MCPManager
 from config.config_loader import get_private_config_from_api
 from config.manage_api_client import DeviceNotFoundException, DeviceBindException
 from core.utils.output_counter import add_device_output
-from core.handle.ttsReportHandle import enqueue_tts_report, report_tts
+from core.handle.reportHandle import enqueue_tts_report, report
 
 TAG = __name__
 
@@ -58,9 +60,9 @@ class ConnectionHandler:
         self.config = copy.deepcopy(config)
         self.session_id = str(uuid.uuid4())
         self.logger = setup_logging()
-        self.auth = AuthMiddleware(config)
         self.server = server  # 保存server实例的引用
 
+        self.auth = AuthMiddleware(config)
         self.need_bind = False
         self.bind_code = None
         self.read_config_from_api = self.config.get("read_config_from_api", False)
@@ -73,6 +75,7 @@ class ConnectionHandler:
         self.prompt = None
         self.welcome_msg = None
         self.max_output_size = 0
+        self.chat_history_conf = 0
 
         # 客户端状态相关
         self.client_abort = False
@@ -86,8 +89,11 @@ class ConnectionHandler:
         self.executor = ThreadPoolExecutor(max_workers=10)
 
         # 上报线程
-        self.tts_report_queue = queue.Queue()
+        self.report_queue = queue.Queue()
-        self.tts_report_thread = None
+        self.report_thread = None
+        # TODO(haotian): 2025/5/12 可以通过修改此处，调节asr的上报和tts的上报
+        self.report_asr_enable = self.read_config_from_api
+        self.report_tts_enable = self.read_config_from_api
 
         # 依赖的组件
         self.vad = None
@@ -128,14 +134,18 @@ class ConnectionHandler:
             if len(cmd) > self.max_cmd_length:
                 self.max_cmd_length = len(cmd)
 
-        self.close_after_chat = False  # 是否在聊天结束后关闭连接
+        # 是否在聊天结束后关闭连接
-        self.use_function_call_mode = False
+        self.close_after_chat = False
+        self.load_function_plugin = False
+        self.intent_type = "nointent"
 
         self.timeout_task = None
         self.timeout_seconds = (
             int(self.config.get("close_connection_no_voice_time", 120)) + 60
         )  # 在原来第一道关闭的基础上加60秒，进行二道关闭
 
+        self.audio_format = "opus"
+
     async def handle_connection(self, ws):
         try:
             # 获取并验证headers
@@ -215,10 +225,26 @@ class ConnectionHandler:
         """保存记忆并关闭连接"""
         try:
             if self.memory:
-                await self.memory.save_memory(self.dialogue.dialogue)
+                # 使用线程池异步保存记忆
+                def save_memory_task():
+                    try:
+                        # 创建新事件循环（避免与主循环冲突）
+                        loop = asyncio.new_event_loop()
+                        asyncio.set_event_loop(loop)
+                        loop.run_until_complete(
+                            self.memory.save_memory(self.dialogue.dialogue)
+                        )
+                    except Exception as e:
+                        self.logger.bind(tag=TAG).error(f"保存记忆失败: {e}")
+                    finally:
+                        loop.close()
+
+                # 启动线程保存记忆，不等待完成
+                threading.Thread(target=save_memory_task, daemon=True).start()
         except Exception as e:
             self.logger.bind(tag=TAG).error(f"保存记忆失败: {e}")
         finally:
+            # 立即关闭连接，不等待记忆保存完成
             await self.close(ws)
 
     async def reset_timeout(self):
@@ -237,11 +263,60 @@ class ConnectionHandler:
         elif isinstance(message, bytes):
             await handleAudioMessage(self, message)
 
+    async def handle_restart(self, message):
+        """处理服务器重启请求"""
+        try:
+
+            self.logger.bind(tag=TAG).info("收到服务器重启指令，准备执行...")
+
+            # 发送确认响应
+            await self.websocket.send(
+                json.dumps(
+                    {
+                        "type": "server_response",
+                        "status": "success",
+                        "message": "服务器重启中...",
+                    }
+                )
+            )
+
+            # 异步执行重启操作
+            def restart_server():
+                """实际执行重启的方法"""
+                time.sleep(1)
+                self.logger.bind(tag=TAG).info("执行服务器重启...")
+                subprocess.Popen(
+                    [sys.executable, "app.py"],
+                    stdin=sys.stdin,
+                    stdout=sys.stdout,
+                    stderr=sys.stderr,
+                    start_new_session=True,
+                )
+                os._exit(0)
+
+            # 使用线程执行重启避免阻塞事件循环
+            threading.Thread(target=restart_server, daemon=True).start()
+
+        except Exception as e:
+            self.logger.bind(tag=TAG).error(f"重启失败: {str(e)}")
+            await self.websocket.send(
+                json.dumps(
+                    {
+                        "type": "server_response",
+                        "status": "error",
+                        "message": f"Restart failed: {str(e)}",
+                    }
+                )
+            )
+
     def _initialize_components(self):
         """初始化组件"""
-        self.prompt = self.config["prompt"]
+        if self.config.get("prompt") is not None:
-        self.change_system_prompt(self.prompt)
+            self.prompt = self.config["prompt"]
-        self.logger.bind(tag=TAG).info(f"初始化组件: prompt成功 {self.prompt[:50]}...")
+            self.change_system_prompt(self.prompt)
+            self.logger.bind(tag=TAG).info(
+                f"初始化组件: prompt成功 {self.prompt[:50]}..."
+            )
 
         """初始化本地组件"""
         if self.vad is None:
@@ -259,11 +334,13 @@ class ConnectionHandler:
         """初始化ASR和TTS上报线程"""
         if not self.read_config_from_api or self.need_bind:
             return
-        if self.tts_report_thread is None or not self.tts_report_thread.is_alive():
+        if self.chat_history_conf == 0:
-            self.tts_report_thread = threading.Thread(
+            return
-                target=self._tts_report_worker, daemon=True
+        if self.report_thread is None or not self.report_thread.is_alive():
+            self.report_thread = threading.Thread(
+                target=self._report_worker, daemon=True
             )
-            self.tts_report_thread.start()
+            self.report_thread.start()
             self.logger.bind(tag=TAG).info("TTS上报线程已启动")
 
     def _initialize_private_config(self):
@@ -330,9 +407,12 @@ class ConnectionHandler:
             ]["Intent"]
         if private_config.get("prompt", None) is not None:
             self.config["prompt"] = private_config["prompt"]
+        if private_config.get("summaryMemory", None) is not None:
+            self.config["summaryMemory"] = private_config["summaryMemory"]
         if private_config.get("device_max_output_size", None) is not None:
             self.max_output_size = int(private_config["device_max_output_size"])
-
+        if private_config.get("chat_history_conf", None) is not None:
+            self.chat_history_conf = int(private_config["chat_history_conf"])
         try:
             modules = initialize_modules(
                 self.logger,
@@ -362,14 +442,19 @@ class ConnectionHandler:
 
     def _initialize_memory(self):
         """初始化记忆模块"""
-        self.memory.init_memory(self.device_id, self.llm)
+        self.memory.init_memory(
+            role_id=self.device_id,
+            llm=self.llm,
+            summary_memory=self.config.get("summaryMemory", None),
+            save_to_file=not self.read_config_from_api,
+        )
 
     def _initialize_intent(self):
-        if (
+        self.intent_type = self.config["Intent"][
-            self.config["Intent"][self.config["selected_module"]["Intent"]]["type"]
+            self.config["selected_module"]["Intent"]
-            == "function_call"
+        ]["type"]
-        ):
+        if self.intent_type == "function_call" or self.intent_type == "intent_llm":
-            self.use_function_call_mode = True
+            self.load_function_plugin = True
         """初始化意图识别模块"""
         # 获取意图识别配置
         intent_config = self.config["Intent"]
@@ -757,7 +842,13 @@ class ConnectionHandler:
                 )
 
                 self.dialogue.put(
-                    Message(role="tool", tool_call_id=function_id, content=text)
+                    Message(
+                        role="tool",
+                        tool_call_id=(
+                            str(uuid.uuid4()) if function_id is None else function_id
+                        ),
+                        content=text,
+                    )
                 )
                 self.chat_with_function_calling(text, tool_call=True)
         elif result.action == Action.NOTFOUND or result.action == Action.ERROR:
@@ -785,7 +876,7 @@ class ConnectionHandler:
                 if future is None:
                     continue
                 text = None
-                opus_datas, tts_file = [], None
+                audio_datas, tts_file = [], None
                 try:
                     self.logger.bind(tag=TAG).debug("正在处理TTS任务...")
                     tts_timeout = int(self.config.get("tts_timeout", 10))
@@ -803,9 +894,12 @@ class ConnectionHandler:
                             f"TTS生成：文件路径: {tts_file}"
                         )
                         if os.path.exists(tts_file):
-                            opus_datas, _ = self.tts.audio_to_opus_data(tts_file)
+                            if self.audio_format == "pcm":
-                            # 在这里上报TTS数据（使用文件路径）
+                                audio_datas, _ = self.tts.audio_to_pcm_data(tts_file)
-                            enqueue_tts_report(self, 2, text, opus_datas)
+                            else:
+                                audio_datas, _ = self.tts.audio_to_opus_data(tts_file)
+                            # 在这里上报TTS数据
+                            enqueue_tts_report(self, text, audio_datas)
                         else:
                             self.logger.bind(tag=TAG).error(
                                 f"TTS出错：文件不存在{tts_file}"
@@ -816,7 +910,7 @@ class ConnectionHandler:
                     self.logger.bind(tag=TAG).error(f"TTS出错: {e}")
                 if not self.client_abort:
                     # 如果没有中途打断就发送语音
-                    self.audio_play_queue.put((opus_datas, text, text_index))
+                    self.audio_play_queue.put((audio_datas, text, text_index))
                 if (
                     self.tts.delete_audio_file
                     and tts_file is not None
@@ -847,13 +941,13 @@ class ConnectionHandler:
             text = None
             try:
                 try:
-                    opus_datas, text, text_index = self.audio_play_queue.get(timeout=1)
+                    audio_datas, text, text_index = self.audio_play_queue.get(timeout=1)
                 except queue.Empty:
                     if self.stop_event.is_set():
                         break
                     continue
                 future = asyncio.run_coroutine_threadsafe(
-                    sendAudioMessage(self, opus_datas, text, text_index), self.loop
+                    sendAudioMessage(self, audio_datas, text, text_index), self.loop
                 )
                 future.result()
             except Exception as e:
@@ -861,13 +955,12 @@ class ConnectionHandler:
                     f"audio_play_priority priority_thread: {text} {e}"
                 )
 
-    def _tts_report_worker(self):
+    def _report_worker(self):
-        """TTS上报工作线程"""
+        """聊天记录上报工作线程"""
-
         while not self.stop_event.is_set():
             try:
                 # 从队列获取数据，设置超时以便定期检查停止事件
-                item = self.tts_report_queue.get(timeout=1)
+                item = self.report_queue.get(timeout=1)
                 if item is None:  # 检测毒丸对象
                     break
 
@@ -875,18 +968,18 @@ class ConnectionHandler:
 
                 try:
                     # 执行上报（传入二进制数据）
-                    report_tts(self, type, text, audio_data)
+                    report(self, type, text, audio_data)
                 except Exception as e:
-                    self.logger.bind(tag=TAG).error(f"TTS上报线程异常: {e}")
+                    self.logger.bind(tag=TAG).error(f"聊天记录上报线程异常: {e}")
                 finally:
                     # 标记任务完成
-                    self.tts_report_queue.task_done()
+                    self.report_queue.task_done()
             except queue.Empty:
                 continue
             except Exception as e:
-                self.logger.bind(tag=TAG).error(f"TTS上报工作线程异常: {e}")
+                self.logger.bind(tag=TAG).error(f"聊天记录上报工作线程异常: {e}")
 
-        self.logger.bind(tag=TAG).info("TTS上报线程已退出")
+        self.logger.bind(tag=TAG).info("聊天记录上报线程已退出")
 
     def speak_and_play(self, text, text_index=0):
         if text is None or len(text) <= 0:
@@ -915,6 +1008,7 @@ class ConnectionHandler:
 
     async def close(self, ws=None):
         """资源清理方法"""
+
         # 取消超时任务
         if self.timeout_task:
             self.timeout_task.cancel()
@@ -924,43 +1018,42 @@ class ConnectionHandler:
         if hasattr(self, "mcp_manager") and self.mcp_manager:
             await self.mcp_manager.cleanup_all()
 
-        # 触发停止事件并清理资源
+        # 触发停止事件
         if self.stop_event:
             self.stop_event.set()
 
-        # 立即关闭线程池
-        if self.executor:
-            self.executor.shutdown(wait=False, cancel_futures=True)
-            self.executor = None
-
-        # 添加毒丸对象到上报队列确保线程退出
-        self.tts_report_queue.put(None)
-
         # 清空任务队列
         self.clear_queues()
 
+        # 关闭WebSocket连接
         if ws:
             await ws.close()
         elif self.websocket:
             await self.websocket.close()
+
+        # 最后关闭线程池（避免阻塞）
+        if self.executor:
+            self.executor.shutdown(wait=False)
+            self.executor = None
+
         self.logger.bind(tag=TAG).info("连接资源已释放")
 
     def clear_queues(self):
-        # 清空所有任务队列
+        """清空所有任务队列"""
         self.logger.bind(tag=TAG).debug(
             f"开始清理: TTS队列大小={self.tts_queue.qsize()}, 音频队列大小={self.audio_play_queue.qsize()}"
         )
+
+        # 使用非阻塞方式清空队列
         for q in [self.tts_queue, self.audio_play_queue]:
             if not q:
                 continue
-            while not q.empty():
+            while True:
                 try:
                     q.get_nowait()
                 except queue.Empty:
-                    continue
+                    break
-            q.queue.clear()
+
-            # 添加毒丸信号到队列，确保线程退出
-            # q.queue.put(None)
         self.logger.bind(tag=TAG).debug(
             f"清理结束: TTS队列大小={self.tts_queue.qsize()}, 音频队列大小={self.audio_play_queue.qsize()}"
         )

main/xiaozhi-server/core/handle/helloHandle.py

@@ -1,5 +1,4 @@
 import json
-from config.logger import setup_logging
 from core.handle.sendAudioHandle import send_stt_message
 from core.utils.util import remove_punctuation_and_length
 import shutil
@@ -20,7 +19,16 @@ WAKEUP_CONFIG = {
 }
 
 
-async def handleHelloMessage(conn):
+async def handleHelloMessage(conn, msg_json):
+    """处理hello消息"""
+    audio_params = msg_json.get("audio_params")
+    if audio_params:
+        format = audio_params.get("format")
+        conn.logger.bind(tag=TAG).info(f"客户端音频格式: {format}")
+        conn.audio_format = format
+        conn.asr.set_audio_format(format)
+        conn.welcome_msg["audio_params"] = audio_params
+
     await conn.websocket.send(json.dumps(conn.welcome_msg))
 
 

main/xiaozhi-server/core/handle/intentHandler.py

@@ -5,6 +5,8 @@ from core.handle.sendAudioHandle import send_stt_message
 from core.handle.helloHandle import checkWakeupWords
 from core.utils.util import remove_punctuation_and_length
 from core.utils.dialogue import Message
+from plugins_func.register import Action
+from loguru import logger
 
 TAG = __name__
 
@@ -17,7 +19,7 @@ async def handle_user_intent(conn, text):
     if await checkWakeupWords(conn, text):
         return True
 
-    if conn.use_function_call_mode:
+    if conn.intent_type == "function_call":
         # 使用支持function calling的聊天方法,不再进行意图分析
         return False
     # 使用LLM进行意图分析
@@ -100,24 +102,35 @@ async def process_intent_result(conn, intent_result, original_text):
                 result = conn.func_handler.handle_llm_function_call(
                     conn, function_call_data
                 )
-                if result and function_name != "play_music":
+                logger.bind(tag=TAG).debug(f"检测到Action : {result.action}")
-                    # 获取当前最新的文本索引
+
-                    text = result.response
+                if result:
-                    if text is None:
+                    if result.action == Action.RESPONSE:  # 直接回复前端
+                        text = result.response
+                        if text is not None:
+                            speak_and_play(conn, text)
+                    elif result.action == Action.REQLLM:  # 调用函数后再请求llm生成回复
                         text = result.result
-                    if text is not None:
+                        conn.dialogue.put(Message(role="tool", content=text))
-                        text_index = (
+                        llm_result = conn.intent.replyResult(text, original_text)
-                            conn.tts_last_text_index + 1
+                        if llm_result is None:
-                            if hasattr(conn, "tts_last_text_index")
+                            llm_result = text
-                            else 0
+                        speak_and_play(conn, llm_result)
-                        )
+                    elif (
-                        conn.recode_first_last_text(text, text_index)
+                        result.action == Action.NOTFOUND
-                        future = conn.executor.submit(
+                        or result.action == Action.ERROR
-                            conn.speak_and_play, text, text_index
+                    ):
-                        )
+                        text = result.result
-                        conn.llm_finish_task = True
+                        if text is not None:
-                        conn.tts_queue.put((future, text_index))
+                            speak_and_play(conn, text)
-                        conn.dialogue.put(Message(role="assistant", content=text))
+                    elif function_name != "play_music":
+                        # For backward compatibility with original code
+                        # 获取当前最新的文本索引
+                        text = result.response
+                        if text is None:
+                            text = result.result
+                        if text is not None:
+                            speak_and_play(conn, text)
 
             # 将函数执行放在线程池中
             conn.executor.submit(process_function_call)
@@ -128,21 +141,12 @@ async def process_intent_result(conn, intent_result, original_text):
         return False
 
 
-def extract_text_in_brackets(s):
+def speak_and_play(conn, text):
-    """
+    text_index = (
-    从字符串中提取中括号内的文字
+        conn.tts_last_text_index + 1 if hasattr(conn, "tts_last_text_index") else 0
-
+    )
-    :param s: 输入字符串
+    conn.recode_first_last_text(text, text_index)
-    :return: 中括号内的文字，如果不存在则返回空字符串
+    future = conn.executor.submit(conn.speak_and_play, text, text_index)
-    """
+    conn.llm_finish_task = True
-    left_bracket_index = s.find("[")
+    conn.tts_queue.put((future, text_index))
-    right_bracket_index = s.find("]")
+    conn.dialogue.put(Message(role="assistant", content=text))
-
-    if (
-        left_bracket_index != -1
-        and right_bracket_index != -1
-        and left_bracket_index < right_bracket_index
-    ):
-        return s[left_bracket_index + 1 : right_bracket_index]
-    else:
-        return ""

main/xiaozhi-server/core/handle/iotHandle.py

@@ -317,7 +317,7 @@ async def handleIotDescriptors(conn, descriptors):
         )
         conn.iot_descriptors[descriptor["name"]] = iot_descriptor
 
-        if conn.use_function_call_mode:
+        if conn.load_function_plugin:
             # 注册或获取设备类型
             type_id = register_device_type(descriptor)
             device_functions = device_type_registry.get_device_functions(type_id)

main/xiaozhi-server/core/handle/receiveAudioHandle.py

@@ -4,8 +4,8 @@ from core.utils.util import remove_punctuation_and_length
 from core.handle.sendAudioHandle import send_stt_message
 from core.handle.intentHandler import handle_user_intent
 from core.utils.output_counter import check_device_output_limit
-from core.handle.ttsReportHandle import enqueue_tts_report
+from core.handle.reportHandle import enqueue_asr_report
-from core.providers.tts.base import audio_to_opus_data
+from core.utils.util import audio_to_data
 
 TAG = __name__
 
@@ -16,7 +16,7 @@ async def handleAudioMessage(conn, audio):
     if not conn.asr_server_receive:
         conn.logger.bind(tag=TAG).debug(f"前期数据处理中，暂停接收")
         return
-    if conn.client_listen_mode == "auto":
+    if conn.client_listen_mode == "auto" or conn.client_listen_mode == "realtime":
         have_voice = conn.vad.is_vad(conn, audio)
     else:
         have_voice = conn.client_have_voice
@@ -44,7 +44,7 @@ async def handleAudioMessage(conn, audio):
             text_len, _ = remove_punctuation_and_length(text)
             if text_len > 0:
                 # 使用自定义模块进行上报
-                enqueue_tts_report(conn, 1, text, copy.deepcopy(conn.asr_audio))
+                enqueue_asr_report(conn, text, copy.deepcopy(conn.asr_audio))
 
                 await startToChat(conn, text)
             else:
@@ -76,7 +76,7 @@ async def startToChat(conn, text):
 
     # 意图未被处理，继续常规聊天流程
     await send_stt_message(conn, text)
-    if conn.use_function_call_mode:
+    if conn.intent_type == "function_call":
         # 使用支持function calling的聊天方法
         conn.executor.submit(conn.chat_with_function_calling, text)
     else:
@@ -98,9 +98,14 @@ async def no_voice_close_connect(conn):
             conn.close_after_chat = True
             conn.client_abort = False
             conn.asr_server_receive = False
-            prompt = (
+            end_prompt = conn.config.get("end_prompt", {})
-                "请你以“时间过得真快”未来头，用富有感情、依依不舍的话来结束这场对话吧。"
+            if end_prompt and end_prompt.get("enable", True) is False:
-            )
+                conn.logger.bind(tag=TAG).info("结束对话，无需发送结束提示语")
+                await conn.close()
+                return
+            prompt = end_prompt.get("prompt")
+            if not prompt:
+                prompt = "请你以“时间过得真快”未来头，用富有感情、依依不舍的话来结束这场对话吧。！"
             await startToChat(conn, prompt)
 
 
@@ -111,7 +116,7 @@ async def max_out_size(conn):
     conn.tts_last_text_index = 0
     conn.llm_finish_task = True
     file_path = "config/assets/max_output_size.wav"
-    opus_packets, _ = audio_to_opus_data(file_path)
+    opus_packets, _ = audio_to_data(file_path)
     conn.audio_play_queue.put((opus_packets, text, 0))
     conn.close_after_chat = True
 
@@ -133,7 +138,7 @@ async def check_bind_device(conn):
 
         # 播放提示音
         music_path = "config/assets/bind_code.wav"
-        opus_packets, _ = audio_to_opus_data(music_path)
+        opus_packets, _ = audio_to_data(music_path)
         conn.audio_play_queue.put((opus_packets, text, 0))
 
         # 逐个播放数字
@@ -141,7 +146,7 @@ async def check_bind_device(conn):
             try:
                 digit = conn.bind_code[i]
                 num_path = f"config/assets/bind_code/{digit}.wav"
-                num_packets, _ = audio_to_opus_data(num_path)
+                num_packets, _ = audio_to_data(num_path)
                 conn.audio_play_queue.put((num_packets, None, i + 1))
             except Exception as e:
                 conn.logger.bind(tag=TAG).error(f"播放数字音频失败: {e}")
@@ -153,5 +158,5 @@ async def check_bind_device(conn):
         conn.tts_last_text_index = 0
         conn.llm_finish_task = True
         music_path = "config/assets/bind_not_found.wav"
-        opus_packets, _ = audio_to_opus_data(music_path)
+        opus_packets, _ = audio_to_data(music_path)
         conn.audio_play_queue.put((opus_packets, text, 0))

main/xiaozhi-server/core/handle/reportHandle.py

@@ -11,13 +11,13 @@ TTS上报功能已集成到ConnectionHandler类中。
 
 import opuslib_next
 
-from config.manage_api_client import report
+from config.manage_api_client import report as manage_report
 
 TAG = __name__
 
 
-def report_tts(conn, type, text, opus_data):
+def report(conn, type, text, opus_data):
-    """执行TTS上报操作
+    """执行聊天记录上报操作
 
     Args:
         conn: 连接对象
@@ -31,7 +31,7 @@ def report_tts(conn, type, text, opus_data):
         else:
             audio_data = None
         # 执行上报
-        report(
+        manage_report(
             mac_address=conn.device_id,
             session_id=conn.session_id,
             chat_type=type,
@@ -39,7 +39,7 @@ def report_tts(conn, type, text, opus_data):
             audio=audio_data,
         )
     except Exception as e:
-        conn.logger.bind(tag=TAG).error(f"TTS上报失败: {e}")
+        conn.logger.bind(tag=TAG).error(f"聊天记录上报失败: {e}")
 
 
 def opus_to_wav(conn, opus_data):
@@ -89,8 +89,10 @@ def opus_to_wav(conn, opus_data):
     return bytes(wav_header) + pcm_data_bytes
 
 
-def enqueue_tts_report(conn, type, text, opus_data):
+def enqueue_tts_report(conn, text, opus_data):
-    if not conn.read_config_from_api or conn.need_bind:
+    if not conn.read_config_from_api or conn.need_bind or not conn.report_tts_enable:
+        return
+    if conn.chat_history_conf == 0:
         return
     """将TTS数据加入上报队列
 
@@ -101,10 +103,43 @@ def enqueue_tts_report(conn, type, text, opus_data):
     """
     try:
         # 使用连接对象的队列，传入文本和二进制数据而非文件路径
-        conn.tts_report_queue.put((type, text, opus_data))
+        if conn.chat_history_conf == 2:
-
+            conn.report_queue.put((2, text, opus_data))
-        conn.logger.bind(tag=TAG).debug(
+            conn.logger.bind(tag=TAG).debug(
-            f"TTS数据已加入上报队列: {conn.device_id}, 音频大小: {len(opus_data)} "
+                f"TTS数据已加入上报队列: {conn.device_id}, 音频大小: {len(opus_data)} "
-        )
+            )
+        else:
+            conn.report_queue.put((2, text, None))
+            conn.logger.bind(tag=TAG).debug(
+                f"TTS数据已加入上报队列: {conn.device_id}, 不上报音频"
+            )
     except Exception as e:
         conn.logger.bind(tag=TAG).error(f"加入TTS上报队列失败: {text}, {e}")
+
+
+def enqueue_asr_report(conn, text, opus_data):
+    if not conn.read_config_from_api or conn.need_bind or not conn.report_asr_enable:
+        return
+    if conn.chat_history_conf == 0:
+        return
+    """将ASR数据加入上报队列
+
+    Args:
+        conn: 连接对象
+        text: 合成文本
+        opus_data: opus音频数据
+    """
+    try:
+        # 使用连接对象的队列，传入文本和二进制数据而非文件路径
+        if conn.chat_history_conf == 2:
+            conn.report_queue.put((1, text, opus_data))
+            conn.logger.bind(tag=TAG).debug(
+                f"ASR数据已加入上报队列: {conn.device_id}, 音频大小: {len(opus_data)} "
+            )
+        else:
+            conn.report_queue.put((1, text, None))
+            conn.logger.bind(tag=TAG).debug(
+                f"ASR数据已加入上报队列: {conn.device_id}, 不上报音频"
+            )
+    except Exception as e:
+        conn.logger.bind(tag=TAG).error(f"加入ASR上报队列失败: {text}, {e}")

main/xiaozhi-server/core/handle/sendAudioHandle.py

@@ -115,7 +115,7 @@ async def send_tts_message(conn, state, text=None):
             stop_tts_notify_voice = conn.config.get(
                 "stop_tts_notify_voice", "config/assets/tts_notify.mp3"
             )
-            audios, duration = conn.tts.audio_to_opus_data(stop_tts_notify_voice)
+            audios, _ = conn.tts.audio_to_opus_data(stop_tts_notify_voice)
             await sendAudio(conn, audios)
         # 清除服务端讲话状态
         conn.clearSpeakStatus()

main/xiaozhi-server/core/handle/textHandle.py

@@ -5,7 +5,7 @@ from core.utils.util import remove_punctuation_and_length
 from core.handle.receiveAudioHandle import startToChat, handleAudioMessage
 from core.handle.sendAudioHandle import send_stt_message, send_tts_message
 from core.handle.iotHandle import handleIotDescriptors, handleIotStatus
-from core.handle.ttsReportHandle import enqueue_tts_report
+from core.handle.reportHandle import enqueue_asr_report
 import asyncio
 
 TAG = __name__
@@ -20,7 +20,7 @@ async def handleTextMessage(conn, message):
             await conn.websocket.send(message)
             return
         if msg_json["type"] == "hello":
-            await handleHelloMessage(conn)
+            await handleHelloMessage(conn, msg_json)
         elif msg_json["type"] == "abort":
             await handleAbortMessage(conn)
         elif msg_json["type"] == "listen":
@@ -56,11 +56,11 @@ async def handleTextMessage(conn, message):
                         await send_tts_message(conn, "stop", None)
                     elif is_wakeup_words:
                         # 上报纯文字数据（复用ASR上报功能，但不提供音频数据）
-                        enqueue_tts_report(conn, 1, "嘿，你好呀", [])
+                        enqueue_asr_report(conn, "嘿，你好呀", [])
                         await startToChat(conn, "嘿，你好呀")
                     else:
                         # 上报纯文字数据（复用ASR上报功能，但不提供音频数据）
-                        enqueue_tts_report(conn, 1, text, [])
+                        enqueue_asr_report(conn, text, [])
                         # 否则需要LLM对文字内容进行答复
                         await startToChat(conn, text)
         elif msg_json["type"] == "iot":
@@ -136,5 +136,8 @@ async def handleTextMessage(conn, message):
                             }
                         )
                     )
+            # 重启服务器
+            elif msg_json["action"] == "restart":
+                await conn.handle_restart(msg_json)
     except json.JSONDecodeError:
         await conn.websocket.send(message)

main/xiaozhi-server/core/providers/asr/aliyun.py

@@ -20,64 +20,78 @@ from core.providers.asr.base import ASRProviderBase
 TAG = __name__
 logger = setup_logging()
 
+
 class AccessToken:
     @staticmethod
     def _encode_text(text):
         encoded_text = parse.quote_plus(text)
-        return encoded_text.replace('+', '%20').replace('*', '%2A').replace('%7E', '~')
+        return encoded_text.replace("+", "%20").replace("*", "%2A").replace("%7E", "~")
 
     @staticmethod
     def _encode_dict(dic):
         keys = dic.keys()
         dic_sorted = [(key, dic[key]) for key in sorted(keys)]
         encoded_text = parse.urlencode(dic_sorted)
-        return encoded_text.replace('+', '%20').replace('*', '%2A').replace('%7E', '~')
+        return encoded_text.replace("+", "%20").replace("*", "%2A").replace("%7E", "~")
 
     @staticmethod
     def create_token(access_key_id, access_key_secret):
-        parameters = {'AccessKeyId': access_key_id,
+        parameters = {
-                      'Action': 'CreateToken',
+            "AccessKeyId": access_key_id,
-                      'Format': 'JSON',
+            "Action": "CreateToken",
-                      'RegionId': 'cn-shanghai',
+            "Format": "JSON",
-                      'SignatureMethod': 'HMAC-SHA1',
+            "RegionId": "cn-shanghai",
-                      'SignatureNonce': str(uuid.uuid1()),
+            "SignatureMethod": "HMAC-SHA1",
-                      'SignatureVersion': '1.0',
+            "SignatureNonce": str(uuid.uuid1()),
-                      'Timestamp': time.strftime("%Y-%m-%dT%H:%M:%SZ", time.gmtime()),
+            "SignatureVersion": "1.0",
-                      'Version': '2019-02-28'}
+            "Timestamp": time.strftime("%Y-%m-%dT%H:%M:%SZ", time.gmtime()),
+            "Version": "2019-02-28",
+        }
         # 构造规范化的请求字符串
         query_string = AccessToken._encode_dict(parameters)
         # print('规范化的请求字符串: %s' % query_string)
         # 构造待签名字符串
-        string_to_sign = 'GET' + '&' + AccessToken._encode_text('/') + '&' + AccessToken._encode_text(query_string)
+        string_to_sign = (
+            "GET"
+            + "&"
+            + AccessToken._encode_text("/")
+            + "&"
+            + AccessToken._encode_text(query_string)
+        )
         # print('待签名的字符串: %s' % string_to_sign)
         # 计算签名
-        secreted_string = hmac.new(bytes(access_key_secret + '&', encoding='utf-8'),
+        secreted_string = hmac.new(
-                                   bytes(string_to_sign, encoding='utf-8'),
+            bytes(access_key_secret + "&", encoding="utf-8"),
-                                   hashlib.sha1).digest()
+            bytes(string_to_sign, encoding="utf-8"),
+            hashlib.sha1,
+        ).digest()
         signature = base64.b64encode(secreted_string)
         # print('签名: %s' % signature)
         # 进行URL编码
         signature = AccessToken._encode_text(signature)
         # print('URL编码后的签名: %s' % signature)
         # 调用服务
-        full_url = 'http://nls-meta.cn-shanghai.aliyuncs.com/?Signature=%s&%s' % (signature, query_string)
+        full_url = "http://nls-meta.cn-shanghai.aliyuncs.com/?Signature=%s&%s" % (
+            signature,
+            query_string,
+        )
         # print('url: %s' % full_url)
         # 提交HTTP GET请求
         response = requests.get(full_url)
         if response.ok:
             root_obj = response.json()
-            key = 'Token'
+            key = "Token"
             if key in root_obj:
-                token = root_obj[key]['Id']
+                token = root_obj[key]["Id"]
-                expire_time = root_obj[key]['ExpireTime']
+                expire_time = root_obj[key]["ExpireTime"]
                 return token, expire_time
         # print(response.text)
         return None, None
 
 
-
 class ASRProvider(ASRProviderBase):
     def __init__(self, config: dict, delete_audio_file: bool):
+        super().__init__()
         """阿里云ASR初始化"""
         # 新增空值判断逻辑
         self.access_key_id = config.get("access_key_id")
@@ -102,28 +116,23 @@ class ASRProvider(ASRProviderBase):
         # 确保输出目录存在
         os.makedirs(self.output_dir, exist_ok=True)
 
-
     def _refresh_token(self):
         """刷新Token并记录过期时间"""
         if self.access_key_id and self.access_key_secret:
             self.token, expire_time_str = AccessToken.create_token(
-                self.access_key_id,
+                self.access_key_id, self.access_key_secret
-                self.access_key_secret
             )
             if not expire_time_str:
                 raise ValueError("无法获取有效的Token过期时间")
 
             try:
-                #统一转换为字符串处理
+                # 统一转换为字符串处理
                 expire_str = str(expire_time_str).strip()
 
                 if expire_str.isdigit():
                     expire_time = datetime.fromtimestamp(int(expire_str))
                 else:
-                    expire_time = datetime.strptime(
+                    expire_time = datetime.strptime(expire_str, "%Y-%m-%dT%H:%M:%SZ")
-                        expire_str,
-                        "%Y-%m-%dT%H:%M:%SZ"
-                    )
                 self.expire_time = expire_time.timestamp() - 60
             except Exception as e:
                 raise ValueError(f"无效的过期时间格式: {expire_str}") from e
@@ -145,9 +154,12 @@ class ASRProvider(ASRProviderBase):
         #              f"过期时间 {datetime.fromtimestamp(self.expire_time)} | "
         #              f"剩余 {remaining:.2f}秒")
         return time.time() > self.expire_time
-    def generate_filename(self, extension=".wav"):
-        return os.path.join(self.output_file, f"tts-{__name__}{datetime.now().date()}@{uuid.uuid4().hex}{extension}")
 
+    def generate_filename(self, extension=".wav"):
+        return os.path.join(
+            self.output_file,
+            f"tts-{__name__}{datetime.now().date()}@{uuid.uuid4().hex}{extension}",
+        )
 
     def _construct_request_url(self) -> str:
         """构造请求URL，包含参数"""
@@ -159,20 +171,6 @@ class ASRProvider(ASRProviderBase):
         request += "&enable_voice_detection=false"
         return request
 
-    def decode_opus(self, opus_data: List[bytes], session_id: str) -> List[bytes]:
-        """将Opus数据解码为PCM"""
-        decoder = opuslib_next.Decoder(16000, 1)  # 16kHz, 单声道
-        pcm_data = []
-
-        for opus_packet in opus_data:
-            try:
-                pcm_frame = decoder.decode(opus_packet, 960)  # 960 samples = 60ms
-                pcm_data.append(pcm_frame)
-            except opuslib_next.OpusError as e:
-                logger.bind(tag=TAG).error(f"Opus解码错误: {e}", exc_info=True)
-
-        return pcm_data
-
     def save_audio_to_file(self, pcm_data: List[bytes], session_id: str) -> str:
         """PCM数据保存为WAV文件"""
         module_name = __name__.split(".")[-1]
@@ -183,7 +181,7 @@ class ASRProvider(ASRProviderBase):
             wf.setnchannels(1)  # 单声道
             wf.setsampwidth(2)  # 16-bit
             wf.setframerate(self.sample_rate)
-            wf.writeframes(b''.join(pcm_data))
+            wf.writeframes(b"".join(pcm_data))
 
         logger.bind(tag=TAG).debug(f"音频文件已保存至: {file_path}")
         return file_path
@@ -193,9 +191,9 @@ class ASRProvider(ASRProviderBase):
         try:
             # 设置HTTP头
             headers = {
-                'X-NLS-Token': self.token,
+                "X-NLS-Token": self.token,
-                'Content-type': 'application/octet-stream',
+                "Content-type": "application/octet-stream",
-                'Content-Length': str(len(pcm_data))
+                "Content-Length": str(len(pcm_data)),
             }
 
             # 创建连接并发送请求
@@ -203,12 +201,12 @@ class ASRProvider(ASRProviderBase):
             request_url = self._construct_request_url()
 
             loop = asyncio.get_event_loop()
-            await loop.run_in_executor(None, lambda: conn.request(
+            await loop.run_in_executor(
-                method='POST',
+                None,
-                url=request_url,
+                lambda: conn.request(
-                body=pcm_data,
+                    method="POST", url=request_url, body=pcm_data, headers=headers
-                headers=headers
+                ),
-            ))
+            )
 
             # 获取响应
             response = await loop.run_in_executor(None, conn.getresponse)
@@ -218,10 +216,10 @@ class ASRProvider(ASRProviderBase):
             # 解析响应
             try:
                 body_json = json.loads(body)
-                status = body_json.get('status')
+                status = body_json.get("status")
 
                 if status == 20000000:
-                    result = body_json.get('result', '')
+                    result = body_json.get("result", "")
                     logger.bind(tag=TAG).debug(f"ASR结果: {result}")
                     return result
                 else:
@@ -236,7 +234,9 @@ class ASRProvider(ASRProviderBase):
             logger.bind(tag=TAG).error(f"ASR请求失败: {e}", exc_info=True)
             return None
 
-    async def speech_to_text(self, opus_data: List[bytes], session_id: str) -> Tuple[Optional[str], Optional[str]]:
+    async def speech_to_text(
+        self, opus_data: List[bytes], session_id: str
+    ) -> Tuple[Optional[str], Optional[str]]:
         """将语音数据转换为文本"""
         if self._is_token_expired():
             logger.warning("Token已过期，正在自动刷新...")
@@ -245,8 +245,11 @@ class ASRProvider(ASRProviderBase):
         file_path = None
         try:
             # 解码Opus为PCM
-            pcm_data = self.decode_opus(opus_data, session_id)
+            if self.audio_format == "pcm":
-            combined_pcm_data = b''.join(pcm_data)
+                pcm_data = opus_data
+            else:
+                pcm_data = self.decode_opus(opus_data)
+            combined_pcm_data = b"".join(pcm_data)
 
             # 判断是否保存为WAV文件
             if self.delete_audio_file:
@@ -264,4 +267,4 @@ class ASRProvider(ASRProviderBase):
 
         except Exception as e:
             logger.bind(tag=TAG).error(f"语音识别失败: {e}", exc_info=True)
-            return "", file_path
+            return "", file_path

main/xiaozhi-server/core/providers/asr/baidu.py

@@ -20,6 +20,7 @@ logger = setup_logging()
 
 class ASRProvider(ASRProviderBase):
     def __init__(self, config: dict, delete_audio_file: bool = True):
+        super().__init__()
         self.app_id = config.get("app_id")
         self.api_key = config.get("api_key")
         self.secret_key = config.get("secret_key")
@@ -49,28 +50,12 @@ class ASRProvider(ASRProviderBase):
 
         return file_path
 
-    @staticmethod
-    def decode_opus(opus_data: List[bytes]) -> bytes:
-        """将Opus音频数据解码为PCM数据"""
-
-        decoder = opuslib_next.Decoder(16000, 1)  # 16kHz, 单声道
-        pcm_data = []
-
-        for opus_packet in opus_data:
-            try:
-                pcm_frame = decoder.decode(opus_packet, 960)  # 960 samples = 60ms
-                pcm_data.append(pcm_frame)
-            except opuslib_next.OpusError as e:
-                logger.bind(tag=TAG).error(f"Opus解码错误: {e}", exc_info=True)
-
-        return pcm_data
-
     async def speech_to_text(
         self, opus_data: List[bytes], session_id: str
     ) -> Tuple[Optional[str], Optional[str]]:
         """将语音数据转换为文本"""
         if not opus_data:
-            logger.bind(tag=TAG).warn("音频数据为空！")
+            logger.bind(tag=TAG).warning("音频数据为空！")
             return None, None
 
         file_path = None
@@ -81,7 +66,10 @@ class ASRProvider(ASRProviderBase):
                 return None, file_path
 
             # 将Opus音频数据解码为PCM
-            pcm_data = self.decode_opus(opus_data)
+            if self.audio_format == "pcm":
+                pcm_data = opus_data
+            else:
+                pcm_data = self.decode_opus(opus_data)
             combined_pcm_data = b"".join(pcm_data)
 
             # 判断是否保存为WAV文件

main/xiaozhi-server/core/providers/asr/base.py

@@ -1,6 +1,6 @@
 from abc import ABC, abstractmethod
 from typing import Optional, Tuple, List
-
+import opuslib_next
 from config.logger import setup_logging
 
 TAG = __name__
@@ -8,12 +8,37 @@ logger = setup_logging()
 
 
 class ASRProviderBase(ABC):
+    def __init__(self):
+        self.audio_format = "opus"
+
     @abstractmethod
     def save_audio_to_file(self, pcm_data: List[bytes], session_id: str) -> str:
         """PCM数据保存为WAV文件"""
         pass
 
     @abstractmethod
-    async def speech_to_text(self, opus_data: List[bytes], session_id: str) -> Tuple[Optional[str], Optional[str]]:
+    async def speech_to_text(
+        self, opus_data: List[bytes], session_id: str
+    ) -> Tuple[Optional[str], Optional[str]]:
         """将语音数据转换为文本"""
         pass
+
+    def set_audio_format(self, format: str) -> None:
+        """设置音频格式"""
+        self.audio_format = format
+
+    @staticmethod
+    def decode_opus(opus_data: List[bytes]) -> bytes:
+        """将Opus音频数据解码为PCM数据"""
+
+        decoder = opuslib_next.Decoder(16000, 1)  # 16kHz, 单声道
+        pcm_data = []
+
+        for opus_packet in opus_data:
+            try:
+                pcm_frame = decoder.decode(opus_packet, 960)  # 960 samples = 60ms
+                pcm_data.append(pcm_frame)
+            except opuslib_next.OpusError as e:
+                logger.bind(tag=TAG).error(f"Opus解码错误: {e}", exc_info=True)
+
+        return pcm_data

main/xiaozhi-server/core/providers/asr/doubao.py

@@ -85,11 +85,12 @@ def parse_response(res):
 
 class ASRProvider(ASRProviderBase):
     def __init__(self, config: dict, delete_audio_file: bool):
+        super().__init__()
         self.appid = config.get("appid")
         self.cluster = config.get("cluster")
         self.access_token = config.get("access_token")
-        self.boosting_table_name = config.get("boosting_table_name")
+        self.boosting_table_name = config.get("boosting_table_name", "")
-        self.correct_table_name = config.get("correct_table_name")
+        self.correct_table_name = config.get("correct_table_name", "")
         self.output_dir = config.get("output_dir")
         self.delete_audio_file = delete_audio_file
 
@@ -226,21 +227,6 @@ class ASRProvider(ASRProviderBase):
             logger.bind(tag=TAG).error(f"ASR request failed: {e}", exc_info=True)
             return None
 
-    @staticmethod
-    def decode_opus(opus_data: List[bytes], session_id: str) -> List[bytes]:
-
-        decoder = opuslib_next.Decoder(16000, 1)  # 16kHz, 单声道
-        pcm_data = []
-
-        for opus_packet in opus_data:
-            try:
-                pcm_frame = decoder.decode(opus_packet, 960)  # 960 samples = 60ms
-                pcm_data.append(pcm_frame)
-            except opuslib_next.OpusError as e:
-                logger.bind(tag=TAG).error(f"Opus解码错误: {e}", exc_info=True)
-
-        return pcm_data
-
     @staticmethod
     def slice_data(data: bytes, chunk_size: int) -> (list, bool):
         """
@@ -265,7 +251,10 @@ class ASRProvider(ASRProviderBase):
         file_path = None
         try:
             # 合并所有opus数据包
-            pcm_data = self.decode_opus(opus_data, session_id)
+            if self.audio_format == "pcm":
+                pcm_data = opus_data
+            else:
+                pcm_data = self.decode_opus(opus_data)
             combined_pcm_data = b"".join(pcm_data)
 
             # 判断是否保存为WAV文件

main/xiaozhi-server/core/providers/asr/fun_local.py

@@ -6,9 +6,7 @@ import io
 from config.logger import setup_logging
 from typing import Optional, Tuple, List
 import uuid
-import opuslib_next
 from core.providers.asr.base import ASRProviderBase
-
 from funasr import AutoModel
 from funasr.utils.postprocess_utils import rich_transcription_postprocess
 
@@ -35,6 +33,7 @@ class CaptureOutput:
 
 class ASRProvider(ASRProviderBase):
     def __init__(self, config: dict, delete_audio_file: bool):
+        super().__init__()
         self.model_dir = config.get("model_dir")
         self.output_dir = config.get("output_dir")  # 修正配置键名
         self.delete_audio_file = delete_audio_file
@@ -46,7 +45,7 @@ class ASRProvider(ASRProviderBase):
                 model=self.model_dir,
                 vad_kwargs={"max_single_segment_time": 30000},
                 disable_update=True,
-                hub="hf"
+                hub="hf",
                 # device="cuda:0",  # 启用GPU加速
             )
 
@@ -64,27 +63,18 @@ class ASRProvider(ASRProviderBase):
 
         return file_path
 
-    @staticmethod
+    async def speech_to_text(
-    def decode_opus(opus_data: List[bytes], session_id: str) -> List[bytes]:
+        self, opus_data: List[bytes], session_id: str
-
+    ) -> Tuple[Optional[str], Optional[str]]:
-        decoder = opuslib_next.Decoder(16000, 1)  # 16kHz, 单声道
-        pcm_data = []
-
-        for opus_packet in opus_data:
-            try:
-                pcm_frame = decoder.decode(opus_packet, 960)  # 960 samples = 60ms
-                pcm_data.append(pcm_frame)
-            except opuslib_next.OpusError as e:
-                logger.bind(tag=TAG).error(f"Opus解码错误: {e}", exc_info=True)
-
-        return pcm_data
-
-    async def speech_to_text(self, opus_data: List[bytes], session_id: str) -> Tuple[Optional[str], Optional[str]]:
         """语音转文本主处理逻辑"""
         file_path = None
         try:
             # 合并所有opus数据包
-            pcm_data = self.decode_opus(opus_data, session_id)
+            if self.audio_format == "pcm":
+                pcm_data = opus_data
+            else:
+                pcm_data = self.decode_opus(opus_data)
+
             combined_pcm_data = b"".join(pcm_data)
 
             # 判断是否保存为WAV文件
@@ -103,7 +93,9 @@ class ASRProvider(ASRProviderBase):
                 batch_size_s=60,
             )
             text = rich_transcription_postprocess(result[0]["text"])
-            logger.bind(tag=TAG).debug(f"语音识别耗时: {time.time() - start_time:.3f}s | 结果: {text}")
+            logger.bind(tag=TAG).debug(
+                f"语音识别耗时: {time.time() - start_time:.3f}s | 结果: {text}"
+            )
 
             return text, file_path
 

main/xiaozhi-server/core/providers/asr/fun_server.py

@@ -9,22 +9,35 @@ import wave
 import websockets
 from config.logger import setup_logging
 import asyncio
+import re
+
 TAG = __name__
 logger = setup_logging()
 
+
 class ASRProvider(ASRProviderBase):
     def __init__(self, config: dict, delete_audio_file: bool):
-        '''
+        """
         Initialize the ASRProvider with server configuration.
         :param config: Dictionary containing 'host', 'port', and 'is_ssl'.
         :param delete_audio_file: Boolean to indicate whether to delete audio files after processing.
-        '''
+        """
-        self.host = config.get('host', 'localhost')
+        super().__init__()
-        self.port = config.get('port', 10095)
+        self.host = config.get("host", "localhost")
-        self.is_ssl = config.get('is_ssl', True)
+        self.port = config.get("port", 10095)
+        self.api_key = config.get("api_key", "none")
+        self.is_ssl = str(config.get("is_ssl", True)).lower() in (
+            "true",
+            "1",
+            "yes",
+        )
         self.output_dir = config.get("output_dir")
         self.delete_audio_file = delete_audio_file
-        self.uri = f"wss://{self.host}:{self.port}" if self.is_ssl else f"ws://{self.host}:{self.port}"
+        self.uri = (
+            f"wss://{self.host}:{self.port}"
+            if self.is_ssl
+            else f"ws://{self.host}:{self.port}"
+        )
         self.ssl_context = ssl.SSLContext() if self.is_ssl else None
         if self.ssl_context:
             self.ssl_context.check_hostname = False
@@ -44,28 +57,11 @@ class ASRProvider(ASRProviderBase):
 
         return file_path
 
-    @staticmethod
-    def decode_opus(opus_data: List[bytes]) -> bytes:
-        """将Opus音频数据解码为PCM数据"""
-        decoder = opuslib_next.Decoder(16000, 1)  # 16kHz, 单声道
-        pcm_data = []
-
-        for opus_packet in opus_data:
-            try:
-                pcm_frame = decoder.decode(opus_packet, 960)  # 960 samples = 60ms
-                pcm_data.append(pcm_frame)
-            except opuslib_next.OpusError as e:
-                logger.bind(tag=TAG).error(f"Opus解码错误: {e}", exc_info=True)
-
-        return pcm_data
-
-
-
     async def _receive_responses(self, ws) -> None:
-        '''
+        """
         Asynchronous generator to receive messages from the WebSocket.
         Yields each message as it is received.
-        '''
+        """
         text = ""
         while True:
             try:
@@ -78,30 +74,35 @@ class ASRProvider(ASRProviderBase):
                 else:
                     text += response_data.get("text", "")
             except asyncio.TimeoutError:
-                logger.bind(tag=TAG).error("Timeout while waiting for response from WebSocket.")
+                logger.bind(tag=TAG).error(
+                    "Timeout while waiting for response from WebSocket."
+                )
                 break
             except websockets.exceptions.ConnectionClosed as e:
                 logger.bind(tag=TAG).error(f"WebSocket connection closed: {e}")
                 break
         return text
+
     async def _send_data(self, ws, pcm_data: bytes, session_id: str) -> tuple:
-        '''
+        """
         Internal method to handle WebSocket communication.
         Reuses the persistent WebSocket connection if available.
         :param pcm_data: PCM audio data to send.
         :param session_id: Unique session identifier.
         :return: Tuple containing recognized text and optional timestamp.
-        '''
+        """
 
         # Send initial configuration message
-        config_message = json.dumps({
+        config_message = json.dumps(
-            "mode": "offline",
+            {
-            "chunk_size": [5, 10, 5],
+                "mode": "offline",
-            "chunk_interval": 10,
+                "chunk_size": [5, 10, 5],
-            "wav_name": session_id,
+                "chunk_interval": 10,
-            "is_speaking": True,
+                "wav_name": session_id,
-            "itn": False
+                "is_speaking": True,
-        })
+                "itn": False,
+            }
+        )
         await ws.send(config_message)
         logger.bind(tag=TAG).debug(f"Sent configuration message: {config_message}")
 
@@ -114,16 +115,20 @@ class ASRProvider(ASRProviderBase):
         await ws.send(end_message)
         logger.bind(tag=TAG).debug(f"Sent end message: {end_message}")
 
-
+    async def speech_to_text(
-    async def speech_to_text(self, opus_data: List[bytes], session_id: str) -> Tuple[Optional[str], Optional[str]]:
+        self, opus_data: List[bytes], session_id: str
-        '''
+    ) -> Tuple[Optional[str], Optional[str]]:
+        """
         Convert speech data to text using FunASR.
         :param opus_data: List of Opus-encoded audio data chunks.
         :param session_id: Unique session identifier.
         :return: Tuple containing recognized text and optional timestamp.
-        '''
+        """
         file_path = None
-        pcm_data = self.decode_opus(opus_data)
+        if self.audio_format == "pcm":
+            pcm_data = opus_data
+        else:
+            pcm_data = self.decode_opus(opus_data)
         combined_pcm_data = b"".join(pcm_data)
 
         # 判断是否保存为WAV文件
@@ -131,17 +136,24 @@ class ASRProvider(ASRProviderBase):
             pass
         else:
             file_path = self.save_audio_to_file(pcm_data, session_id)
-
+        auth_header = {"Authorization": "Bearer; {}".format(self.api_key)}
-        async with websockets.connect(self.uri, subprotocols=["binary"], ping_interval=None, ssl=self.ssl_context) as ws:
+        async with websockets.connect(
+            self.uri,
+            additional_headers=auth_header,
+            subprotocols=["binary"],
+            ping_interval=None,
+            ssl=self.ssl_context,
+        ) as ws:
             try:
                 # Use asyncio to handle WebSocket communication
-                send_task = asyncio.create_task(self._send_data(ws, combined_pcm_data, session_id))
+                send_task = asyncio.create_task(
+                    self._send_data(ws, combined_pcm_data, session_id)
+                )
                 receive_task = asyncio.create_task(self._receive_responses(ws))
 
                 # Gather tasks with error handling
                 done, pending = await asyncio.wait(
-                    [send_task, receive_task],
+                    [send_task, receive_task], return_when=asyncio.FIRST_EXCEPTION
-                    return_when=asyncio.FIRST_EXCEPTION
                 )
 
                 # Cancel any pending tasks
@@ -155,11 +167,19 @@ class ASRProvider(ASRProviderBase):
 
                 # Get the result from the receive task
                 result = receive_task.result()
-                return result, file_path  # Return the recognized text and timestamp (if any)
+                match = re.match(r"<\|(.*?)\|><\|(.*?)\|><\|(.*?)\|>(.*)", result)
+                if match:
+                    result = match.group(4).strip()
+                return (
+                    result,
+                    file_path,
+                )  # Return the recognized text and timestamp (if any)
 
             except websockets.exceptions.ConnectionClosed as e:
                 logger.bind(tag=TAG).error(f"WebSocket connection closed: {e}")
                 return "", file_path
             except Exception as e:
-                logger.bind(tag=TAG).error(f"Error during speech-to-text conversion: {e}", exc_info=True)
+                logger.bind(tag=TAG).error(
-                return "", file_path
+                    f"Error during speech-to-text conversion: {e}", exc_info=True
+                )
+                return "", file_path

main/xiaozhi-server/core/providers/asr/sherpa_onnx_local.py

@@ -37,6 +37,7 @@ class CaptureOutput:
 
 class ASRProvider(ASRProviderBase):
     def __init__(self, config: dict, delete_audio_file: bool):
+        super().__init__()
         self.model_dir = config.get("model_dir")
         self.output_dir = config.get("output_dir")
         self.delete_audio_file = delete_audio_file
@@ -97,21 +98,6 @@ class ASRProvider(ASRProviderBase):
 
         return file_path
 
-    @staticmethod
-    def decode_opus(opus_data: List[bytes], session_id: str) -> List[bytes]:
-
-        decoder = opuslib_next.Decoder(16000, 1)  # 16kHz, 单声道
-        pcm_data = []
-
-        for opus_packet in opus_data:
-            try:
-                pcm_frame = decoder.decode(opus_packet, 960)  # 960 samples = 60ms
-                pcm_data.append(pcm_frame)
-            except opuslib_next.OpusError as e:
-                logger.bind(tag=TAG).error(f"Opus解码错误: {e}", exc_info=True)
-
-        return pcm_data
-
     def read_wave(self, wave_filename: str) -> Tuple[np.ndarray, int]:
         """
         Args:
@@ -144,7 +130,10 @@ class ASRProvider(ASRProviderBase):
         try:
             # 保存音频文件
             start_time = time.time()
-            pcm_data = self.decode_opus(opus_data, session_id)
+            if self.audio_format == "pcm":
+                pcm_data = opus_data
+            else:
+                pcm_data = self.decode_opus(opus_data)
             file_path = self.save_audio_to_file(pcm_data, session_id)
             logger.bind(tag=TAG).debug(
                 f"音频文件保存耗时: {time.time() - start_time:.3f}s | 路径: {file_path}"

main/xiaozhi-server/core/providers/asr/tencent.py

@@ -17,12 +17,14 @@ from config.logger import setup_logging
 TAG = __name__
 logger = setup_logging()
 
+
 class ASRProvider(ASRProviderBase):
     API_URL = "https://asr.tencentcloudapi.com"
     API_VERSION = "2019-06-14"
     FORMAT = "pcm"  # 支持的音频格式：pcm, wav, mp3
 
     def __init__(self, config: dict, delete_audio_file: bool = True):
+        super().__init__()
         self.secret_id = config.get("secret_id")
         self.secret_key = config.get("secret_key")
         self.output_dir = config.get("output_dir")
@@ -45,26 +47,12 @@ class ASRProvider(ASRProviderBase):
 
         return file_path
 
-    @staticmethod
+    async def speech_to_text(
-    def decode_opus(opus_data: List[bytes]) -> bytes:
+        self, opus_data: List[bytes], session_id: str
-        """将Opus音频数据解码为PCM数据"""
+    ) -> Tuple[Optional[str], Optional[str]]:
-
-        decoder = opuslib_next.Decoder(16000, 1)  # 16kHz, 单声道
-        pcm_data = []
-
-        for opus_packet in opus_data:
-            try:
-                pcm_frame = decoder.decode(opus_packet, 960)  # 960 samples = 60ms
-                pcm_data.append(pcm_frame)
-            except opuslib_next.OpusError as e:
-                logger.bind(tag=TAG).error(f"Opus解码错误: {e}", exc_info=True)
-
-        return pcm_data
-
-    async def speech_to_text(self, opus_data: List[bytes], session_id: str) -> Tuple[Optional[str], Optional[str]]:
         """将语音数据转换为文本"""
         if not opus_data:
-            logger.bind(tag=TAG).warn("音频数据为空！")
+            logger.bind(tag=TAG).warning("音频数据为空！")
             return None, None
 
         file_path = None
@@ -75,7 +63,10 @@ class ASRProvider(ASRProviderBase):
                 return None, file_path
 
             # 将Opus音频数据解码为PCM
-            pcm_data = self.decode_opus(opus_data)
+            if self.audio_format == "pcm":
+                pcm_data = opus_data
+            else:
+                pcm_data = self.decode_opus(opus_data)
             combined_pcm_data = b"".join(pcm_data)
 
             # 判断是否保存为WAV文件
@@ -85,7 +76,7 @@ class ASRProvider(ASRProviderBase):
                 self.save_audio_to_file(pcm_data, session_id)
 
             # 将音频数据转换为Base64编码
-            base64_audio = base64.b64encode(combined_pcm_data).decode('utf-8')
+            base64_audio = base64.b64encode(combined_pcm_data).decode("utf-8")
 
             # 构建请求体
             request_body = self._build_request_body(base64_audio)
@@ -98,7 +89,9 @@ class ASRProvider(ASRProviderBase):
             result = self._send_request(request_body, timestamp, authorization)
 
             if result:
-                logger.bind(tag=TAG).debug(f"腾讯云语音识别耗时: {time.time() - start_time:.3f}s | 结果: {result}")
+                logger.bind(tag=TAG).debug(
+                    f"腾讯云语音识别耗时: {time.time() - start_time:.3f}s | 结果: {result}"
+                )
 
             return result, file_path
 
@@ -115,7 +108,7 @@ class ASRProvider(ASRProviderBase):
             "SourceType": 1,  # 音频数据来源为语音文件
             "VoiceFormat": self.FORMAT,  # 音频格式
             "Data": base64_audio,  # Base64编码的音频数据
-            "DataLen": len(base64_audio)  # 数据长度
+            "DataLen": len(base64_audio),  # 数据长度
         }
         return json.dumps(request_map)
 
@@ -148,9 +141,11 @@ class ASRProvider(ASRProviderBase):
             action = "SentenceRecognition"  # 接口名称
 
             # 构建规范头部信息，注意顺序和格式
-            canonical_headers = f"content-type:{content_type.lower()}\n" + \
+            canonical_headers = (
-                               f"host:{host.lower()}\n" + \
+                f"content-type:{content_type.lower()}\n"
-                               f"x-tc-action:{action.lower()}\n"
+                + f"host:{host.lower()}\n"
+                + f"x-tc-action:{action.lower()}\n"
+            )
 
             signed_headers = "content-type;host;x-tc-action"
 
@@ -158,21 +153,25 @@ class ASRProvider(ASRProviderBase):
             payload_hash = self._sha256_hex(request_body)
 
             # 构建规范请求字符串
-            canonical_request = f"{http_request_method}\n" + \
+            canonical_request = (
-                               f"{canonical_uri}\n" + \
+                f"{http_request_method}\n"
-                               f"{canonical_query_string}\n" + \
+                + f"{canonical_uri}\n"
-                               f"{canonical_headers}\n" + \
+                + f"{canonical_query_string}\n"
-                               f"{signed_headers}\n" + \
+                + f"{canonical_headers}\n"
-                               f"{payload_hash}"
+                + f"{signed_headers}\n"
+                + f"{payload_hash}"
+            )
 
             # 计算规范请求的哈希值
             hashed_canonical_request = self._sha256_hex(canonical_request)
 
             # 构建待签名字符串
-            string_to_sign = f"{algorithm}\n" + \
+            string_to_sign = (
-                            f"{timestamp}\n" + \
+                f"{algorithm}\n"
-                            f"{credential_scope}\n" + \
+                + f"{timestamp}\n"
-                            f"{hashed_canonical_request}"
+                + f"{credential_scope}\n"
+                + f"{hashed_canonical_request}"
+            )
 
             # 计算签名密钥
             secret_date = self._hmac_sha256(f"TC3{self.secret_key}", date)
@@ -180,13 +179,17 @@ class ASRProvider(ASRProviderBase):
             secret_signing = self._hmac_sha256(secret_service, "tc3_request")
 
             # 计算签名
-            signature = self._bytes_to_hex(self._hmac_sha256(secret_signing, string_to_sign))
+            signature = self._bytes_to_hex(
+                self._hmac_sha256(secret_signing, string_to_sign)
+            )
 
             # 构建授权头
-            authorization = f"{algorithm} " + \
+            authorization = (
-                           f"Credential={self.secret_id}/{credential_scope}, " + \
+                f"{algorithm} "
-                           f"SignedHeaders={signed_headers}, " + \
+                + f"Credential={self.secret_id}/{credential_scope}, "
-                           f"Signature={signature}"
+                + f"SignedHeaders={signed_headers}, "
+                + f"Signature={signature}"
+            )
 
             return timestamp, authorization
 
@@ -194,7 +197,9 @@ class ASRProvider(ASRProviderBase):
             logger.bind(tag=TAG).error(f"生成认证头失败: {e}", exc_info=True)
             raise RuntimeError(f"生成认证头失败: {e}")
 
-    def _send_request(self, request_body: str, timestamp: str, authorization: str) -> Optional[str]:
+    def _send_request(
+        self, request_body: str, timestamp: str, authorization: str
+    ) -> Optional[str]:
         """发送请求到腾讯云API"""
         headers = {
             "Content-Type": "application/json; charset=utf-8",
@@ -203,7 +208,7 @@ class ASRProvider(ASRProviderBase):
             "X-TC-Action": "SentenceRecognition",
             "X-TC-Version": self.API_VERSION,
             "X-TC-Timestamp": timestamp,
-            "X-TC-Region": "ap-shanghai"
+            "X-TC-Region": "ap-shanghai",
         }
 
         try:
@@ -225,7 +230,7 @@ class ASRProvider(ASRProviderBase):
             if "Response" in response_json and "Result" in response_json["Response"]:
                 return response_json["Response"]["Result"]
             else:
-                logger.bind(tag=TAG).warn(f"响应中没有识别结果: {response_json}")
+                logger.bind(tag=TAG).warning(f"响应中没有识别结果: {response_json}")
                 return ""
 
         except Exception as e:
@@ -234,16 +239,16 @@ class ASRProvider(ASRProviderBase):
 
     def _sha256_hex(self, data: str) -> str:
         """计算字符串的SHA256哈希值"""
-        digest = hashlib.sha256(data.encode('utf-8')).digest()
+        digest = hashlib.sha256(data.encode("utf-8")).digest()
         return self._bytes_to_hex(digest)
 
     def _hmac_sha256(self, key, data: str) -> bytes:
         """计算HMAC-SHA256"""
         if isinstance(key, str):
-            key = key.encode('utf-8')
+            key = key.encode("utf-8")
 
-        return hmac.new(key, data.encode('utf-8'), hashlib.sha256).digest()
+        return hmac.new(key, data.encode("utf-8"), hashlib.sha256).digest()
 
     def _bytes_to_hex(self, bytes_data: bytes) -> str:
         """字节数组转十六进制字符串"""
-        return ''.join(f"{b:02x}" for b in bytes_data)
+        return "".join(f"{b:02x}" for b in bytes_data)

main/xiaozhi-server/core/providers/intent/base.py

@@ -9,18 +9,6 @@ logger = setup_logging()
 class IntentProviderBase(ABC):
     def __init__(self, config):
         self.config = config
-        self.intent_options = [
-            {
-                "name": "handle_exit_intent",
-                "desc": "结束聊天, 用户发来如再见之类的表示结束的话, 不想再进行对话的时候",
-            },
-            {
-                "name": "play_music",
-                "desc": "播放音乐, 用户希望你可以播放音乐, 只用于播放音乐的意图",
-            },
-            {"name": "get_time", "desc": "获取今天日期或者当前时间信息"},
-            {"name": "continue_chat", "desc": "继续聊天"},
-        ]
 
     def set_llm(self, llm):
         self.llm = llm

main/xiaozhi-server/core/providers/intent/intent_llm/intent_llm.py

@@ -15,57 +15,72 @@ class IntentProvider(IntentProviderBase):
     def __init__(self, config):
         super().__init__(config)
         self.llm = None
-        self.promot = self.get_intent_system_prompt()
+        self.promot = ""
         # 添加缓存管理
         self.intent_cache = {}  # 缓存意图识别结果
         self.cache_expiry = 600  # 缓存有效期10分钟
         self.cache_max_size = 100  # 最多缓存100个意图
+        self.history_count = 4  # 默认使用最近4条对话记录
 
-    def get_intent_system_prompt(self) -> str:
+    def get_intent_system_prompt(self, functions_list: str) -> str:
         """
-        根据配置的意图选项动态生成系统提示词
+        根据配置的意图选项和可用函数动态生成系统提示词
+        Args:
+            functions: 可用的函数列表，JSON格式字符串
         Returns:
             格式化后的系统提示词
         """
 
+        # 构建函数说明部分
+        functions_desc = "可用的函数列表：\n"
+        for func in functions_list:
+            func_info = func.get("function", {})
+            name = func_info.get("name", "")
+            desc = func_info.get("description", "")
+            params = func_info.get("parameters", {})
+
+            functions_desc += f"\n函数名: {name}\n"
+            functions_desc += f"描述: {desc}\n"
+
+            if params:
+                functions_desc += "参数:\n"
+                for param_name, param_info in params.get("properties", {}).items():
+                    param_desc = param_info.get("description", "")
+                    param_type = param_info.get("type", "")
+                    functions_desc += f"- {param_name} ({param_type}): {param_desc}\n"
+
+            functions_desc += "---\n"
+
         prompt = (
-            "你是一个意图识别助手。请分析用户的最后一句话，判断用户意图属于以下哪一类：\n"
+            "你是一个意图识别助手。请分析用户的最后一句话，判断用户意图并调用相应的函数。\n\n"
-            "<start>"
+            f"{functions_desc}\n"
-            f"{str(self.intent_options)}"
+            "处理步骤:\n"
-            "<end>\n"
+            "1. 分析用户输入，确定用户意图\n"
-            "处理步骤:"
+            "2. 从可用函数列表中选择最匹配的函数\n"
-            "1. 思考意图类型，生成function_call格式"
+            "3. 如果找到匹配的函数，生成对应的function_call 格式\n"
-            "\n\n"
+            '4. 如果没有找到匹配的函数，返回{"function_call": {"name": "continue_chat"}}\n\n'
-            "返回格式示例：\n"
+            "返回格式要求：\n"
-            '1. 播放音乐意图: {"function_call": {"name": "play_music", "arguments": {"song_name": "音乐名称"}}}\n'
+            "1. 必须返回纯JSON格式\n"
-            '2. 结束对话意图: {"function_call": {"name": "handle_exit_intent", "arguments": {"say_goodbye": "goodbye"}}}\n'
+            "2. 必须包含function_call字段\n"
-            '3. 获取当天日期时间: {"function_call": {"name": "get_time"}}\n'
+            "3. function_call必须包含name字段\n"
-            '4. 继续聊天意图: {"function_call": {"name": "continue_chat"}}\n'
+            "4. 如果函数需要参数，必须包含arguments字段\n\n"
-            "\n"
+            "示例：\n"
-            "注意:\n"
-            '- 播放音乐：无歌名时，song_name设为"random"\n'
-            "- 如果没有明显的意图，应按照继续聊天意图处理\n"
-            "- 只返回纯JSON，不要任何其他内容\n"
-            "\n"
-            "示例分析:\n"
             "```\n"
-            "用户: 你也太搞笑了\n"
+            "用户: 现在几点了？\n"
-            '返回: {"function_call": {"name": "continue_chat"}}\n'
-            "```\n"
-            "```\n"
-            "用户: 现在是几号了?现在几点了？\n"
             '返回: {"function_call": {"name": "get_time"}}\n'
             "```\n"
             "```\n"
-            "用户: 我们明天再聊吧\n"
+            "用户: 我想结束对话\n"
-            '返回: {"function_call": {"name": "handle_exit_intent"}}\n'
+            '返回: {"function_call": {"name": "handle_exit_intent", "arguments": {"say_goodbye": "goodbye"}}}\n'
             "```\n"
             "```\n"
-            "用户: 播放中秋月\n"
+            "用户: 你好啊\n"
-            '返回: {"function_call": {"name": "play_music", "arguments": {"song_name": "中秋月"}}}\n'
+            '返回: {"function_call": {"name": "continue_chat"}}\n'
-            "```\n"
+            "```\n\n"
-            "```\n"
+            "注意：\n"
-            "可用的音乐名称:\n"
+            "1. 只返回JSON格式，不要包含任何其他文字\n"
+            '2. 如果没有找到匹配的函数，返回{"function_call": {"name": "continue_chat"}}\n'
+            "3. 确保返回的JSON格式正确，包含所有必要的字段\n"
         )
         return prompt
 
@@ -90,6 +105,14 @@ class IntentProvider(IntentProviderBase):
             for key, _ in sorted_items[: len(sorted_items) - self.cache_max_size]:
                 del self.intent_cache[key]
 
+    def replyResult(self, text: str, original_text: str):
+        llm_result = self.llm.response_no_stream(
+            system_prompt=text,
+            user_prompt="请根据以上内容，像人类一样说话的口吻回复用户，要求简洁，请直接返回结果。用户现在说："
+            + original_text,
+        )
+        return llm_result
+
     async def detect_intent(self, conn, dialogue_history: List[Dict], text: str) -> str:
         if not self.llm:
             raise ValueError("LLM provider not set")
@@ -118,22 +141,35 @@ class IntentProvider(IntentProviderBase):
         # 清理缓存
         self.clean_cache()
 
-        # 构建用户最后一句话的提示
+        if self.promot == "":
-        msgStr = ""
+            if hasattr(conn, "func_handler"):
+                functions = conn.func_handler.get_functions()
+                self.promot = self.get_intent_system_prompt(functions)
 
-        # 只使用最后两句即可
-        if len(dialogue_history) >= 2:
-            # 保证最少有两句话的时候处理
-            msgStr += f"{dialogue_history[-2].role}: {dialogue_history[-2].content}\n"
-        msgStr += f"{dialogue_history[-1].role}: {dialogue_history[-1].content}\n"
-
-        msgStr += f"User: {text}\n"
-        user_prompt = f"当前的对话如下：\n{msgStr}"
         music_config = initialize_music_handler(conn)
         music_file_names = music_config["music_file_names"]
-        prompt_music = f"{self.promot}\n<start>{music_file_names}\n<end>"
+        prompt_music = f"{self.promot}\n<musicNames>{music_file_names}\n</musicNames>"
+
+        devices = conn.config["plugins"]["home_assistant"].get("devices", [])
+        if len(devices) > 0:
+            hass_prompt = "\n下面是我家智能设备列表（位置，设备名，entity_id），可以通过homeassistant控制\n"
+            for device in devices:
+                hass_prompt += device + "\n"
+            prompt_music += hass_prompt
+
         logger.bind(tag=TAG).debug(f"User prompt: {prompt_music}")
 
+        # 构建用户对话历史的提示
+        msgStr = ""
+
+        # 获取最近的对话历史
+        start_idx = max(0, len(dialogue_history) - self.history_count)
+        for i in range(start_idx, len(dialogue_history)):
+            msgStr += f"{dialogue_history[i].role}: {dialogue_history[i].content}\n"
+
+        msgStr += f"User: {text}\n"
+        user_prompt = f"current dialogue:\n{msgStr}"
+
         # 记录预处理完成时间
         preprocess_time = time.time() - total_start_time
         logger.bind(tag=TAG).debug(f"意图识别预处理耗时: {preprocess_time:.4f}秒")
@@ -179,7 +215,7 @@ class IntentProvider(IntentProviderBase):
 
                 # 记录识别到的function call
                 logger.bind(tag=TAG).info(
-                    f"识别到function call: {function_name}, 参数: {function_args}"
+                    f"llm 识别到意图: {function_name}, 参数: {function_args}"
                 )
 
                 # 添加到缓存

main/xiaozhi-server/core/providers/llm/AliBL/AliBL.py

@@ -61,5 +61,6 @@ class LLMProvider(LLMProviderBase):
             yield "【LLM服务响应异常】"
 
     def response_with_functions(self, session_id, dialogue, functions=None):
-        logger.bind(tag=TAG).info(f"阿里百练暂未实现完整的工具调用（function call）")
+        logger.bind(tag=TAG).error(
-        return self.response(session_id, dialogue)
+            f"阿里百练暂未实现完整的工具调用（function call），建议使用其他意图识别"
+        )

main/xiaozhi-server/core/providers/llm/fastgpt/fastgpt.py

@@ -66,5 +66,6 @@ class LLMProvider(LLMProviderBase):
             yield "【服务响应异常】"
 
     def response_with_functions(self, session_id, dialogue, functions=None):
-        logger.bind(tag=TAG).info(f"fastgpt暂未实现完整的工具调用（function call）")
+        logger.bind(tag=TAG).error(
-        return self.response(session_id, dialogue)
+            f"fastgpt暂未实现完整的工具调用（function call），建议使用其他意图识别"
+        )

main/xiaozhi-server/core/providers/llm/gemini/gemini.py

@@ -1,140 +1,205 @@
-import google.generativeai as genai
+import os, json, uuid
-from core.utils.util import check_model_key
+from types import SimpleNamespace
-from core.providers.llm.base import LLMProviderBase
+from typing import Any, Dict, List
-from config.logger import setup_logging
-import requests
-import json
 
+import requests
+from google import generativeai as genai
+from google.generativeai import types, GenerationConfig
+
+from core.providers.llm.base import LLMProviderBase
+from core.utils.util import check_model_key
+from config.logger import setup_logging
+from google.generativeai.types import GenerateContentResponse
+from requests import RequestException
+
+log = setup_logging()
 TAG = __name__
-logger = setup_logging()
+
+
+def test_proxy(proxy_url: str, test_url: str) -> bool:
+    try:
+        resp = requests.get(test_url, proxies={"http": proxy_url, "https": proxy_url})
+        return 200 <= resp.status_code < 400
+    except RequestException:
+        return False
+
+
+def setup_proxy_env(http_proxy: str | None, https_proxy: str | None):
+    """
+    分别测试 HTTP 和 HTTPS 代理是否可用，并设置环境变量。
+    如果 HTTPS 代理不可用但 HTTP 可用，会将 HTTPS_PROXY 也指向 HTTP。
+    """
+    test_http_url = "http://www.google.com"
+    test_https_url = "https://www.google.com"
+
+    ok_http = ok_https = False
+
+    if http_proxy:
+        ok_http = test_proxy(http_proxy, test_http_url)
+        if ok_http:
+            os.environ["HTTP_PROXY"] = http_proxy
+            log.bind(tag=TAG).info(f"配置提供的Gemini HTTPS代理连通成功: {http_proxy}")
+        else:
+            log.bind(tag=TAG).warning(f"配置提供的Gemini HTTP代理不可用: {http_proxy}")
+
+    if https_proxy:
+        ok_https = test_proxy(https_proxy, test_https_url)
+        if ok_https:
+            os.environ["HTTPS_PROXY"] = https_proxy
+            log.bind(tag=TAG).info(f"配置提供的Gemini HTTPS代理连通成功: {https_proxy}")
+        else:
+            log.bind(tag=TAG).warning(
+                f"配置提供的Gemini HTTPS代理不可用: {https_proxy}"
+            )
+
+    # 如果https_proxy不可用，但http_proxy可用且能走通https，则复用http_proxy作为https_proxy
+    if ok_http and not ok_https:
+        if test_proxy(http_proxy, test_https_url):
+            os.environ["HTTPS_PROXY"] = http_proxy
+            ok_https = True
+            log.bind(tag=TAG).info(f"复用HTTP代理作为HTTPS代理: {http_proxy}")
+
+    if not ok_http and not ok_https:
+        log.bind(tag=TAG).error(
+            f"Gemini 代理设置失败: HTTP 和 HTTPS 代理都不可用，请检查配置"
+        )
+        raise RuntimeError("HTTP 和 HTTPS 代理都不可用，请检查配置")
 
 
 class LLMProvider(LLMProviderBase):
-    def __init__(self, config):
+    def __init__(self, cfg: Dict[str, Any]):
-        """初始化Gemini LLM Provider"""
+        self.model_name = cfg.get("model_name", "gemini-2.0-flash")
-        self.model_name = config.get("model_name", "gemini-1.5-pro")
+        self.api_key = cfg["api_key"]
-        self.api_key = config.get("api_key")
+        http_proxy = cfg.get("http_proxy")
-        self.http_proxy = config.get("http_proxy")
+        https_proxy = cfg.get("https_proxy")
-        self.https_proxy = config.get("https_proxy")
-        have_key = check_model_key("LLM", self.api_key)
 
-        if not have_key:
+        if not check_model_key("LLM", self.api_key):
-            return
+            raise ValueError("无效的Gemini API Key，请检查是否配置正确")
 
-        try:
+        if http_proxy or https_proxy:
-            # 初始化Gemini客户端
+            log.bind(tag=TAG).info(
-            # 配置代理（如果提供了代理配置）
+                f"检测到Gemini代理配置，开始测试代理连通性和设置代理环境..."
-            self.proxies = None
+            )
-            if self.http_proxy is not "" or self.https_proxy is not "":
+            setup_proxy_env(http_proxy, https_proxy)
+            log.bind(tag=TAG).info(
+                f"Gemini 代理设置成功 - HTTP: {http_proxy}, HTTPS: {https_proxy}"
+            )
+        genai.configure(api_key=self.api_key)
+        self.model = genai.GenerativeModel(self.model_name)
 
-                self.proxies = {
+        self.gen_cfg = GenerationConfig(
-                    "http": self.http_proxy,
+            temperature=0.7,
-                    "https": self.https_proxy,
+            top_p=0.9,
-                }
+            top_k=40,
-                logger.bind(tag=TAG).info(f"Gemini set proxys:{self.proxies}")
+            max_output_tokens=2048,
-                # 使用猴子补丁修改 google-generativeai 库的请求会话
+        )
 
-                # 使用 session 对象配置 genai
+    @staticmethod
-
+    def _build_tools(funcs: List[Dict[str, Any]] | None):
-            genai.configure(api_key=self.api_key)
+        if not funcs:
-            self.model = genai.GenerativeModel(self.model_name)
+            return None
-
+        return [
-            # 设置生成参数
+            types.Tool(
-            self.generation_config = {
+                function_declarations=[
-                "temperature": 0.7,
+                    types.FunctionDeclaration(
-                "top_p": 0.9,
+                        name=f["function"]["name"],
-                "top_k": 40,
+                        description=f["function"]["description"],
-                "max_output_tokens": 2048,
+                        parameters=f["function"]["parameters"],
-            }
+                    )
-            self.chat = None
+                    for f in funcs
-        except Exception as e:
+                ]
-            logger.bind(tag=TAG).error(f"Gemini初始化失败: {e}")
+            )
-            self.model = None
+        ]
 
+    # Gemini文档提到，无需维护session-id，直接用dialogue拼接而成
     def response(self, session_id, dialogue):
-        """生成Gemini对话响应"""
+        yield from self._generate(dialogue, None)
-        if not self.model:
-            yield "【Gemini服务未正确初始化】"
-            return
-
-        try:
-            # 处理对话历史
-            chat_history = []
-            for msg in dialogue[:-1]:  # 历史对话
-                role = "model" if msg["role"] == "assistant" else "user"
-                content = msg["content"].strip()
-                if content:
-                    chat_history.append({"role": role, "parts": [{"text": content}]})
-
-            # 获取当前消息
-            current_msg = dialogue[-1]["content"]
-
-            # 构建请求体
-            request_body = {
-                "contents": chat_history
-                + [{"role": "user", "parts": [{"text": current_msg}]}],
-                "generationConfig": self.generation_config,
-            }
-
-            # 构建请求URL
-            url = f"https://generativelanguage.googleapis.com/v1beta/models/{self.model_name}:generateContent?key={self.api_key}"
-
-            # 构建请求头
-            headers = {
-                "Content-Type": "application/json",
-            }
-
-            # 发送POST请求,经测试手动 request 无法使用 stream 模式
-            if self.proxies:
-                response = requests.post(
-                    url,
-                    headers=headers,
-                    json=request_body,
-                    stream=False,
-                    proxies=self.proxies,
-                )
-                try:
-                    data = response.json()  # 直接解析JSON
-                    if "candidates" in data and data["candidates"]:
-                        yield data["candidates"][0]["content"]["parts"][0]["text"]
-                    else:
-                        yield "未找到候选回复。"
-                except json.JSONDecodeError as e:
-                    yield f"JSON解码错误：{e}"
-                except Exception as e:
-                    yield f"发生错误：{e}"
-            else:
-                logger.bind(tag=TAG).info(f"Gemini stream mode ")
-                chat = self.model.start_chat(history=chat_history)
-
-                # 发送消息并获取流式响应
-                response = chat.send_message(
-                    current_msg, stream=True, generation_config=self.generation_config
-                )
-                # 处理流式响应
-                for chunk in response:
-                    if hasattr(chunk, "text") and chunk.text:
-                        yield chunk.text
-
-        except Exception as e:
-            error_msg = str(e)
-            logger.bind(tag=TAG).error(f"Gemini响应生成错误: {error_msg}")
-
-            # 针对不同错误返回友好提示
-            if "Rate limit" in error_msg:
-                yield "【Gemini服务请求太频繁,请稍后再试】"
-            elif "Invalid API key" in error_msg:
-                yield "【Gemini API key无效】"
-            else:
-                yield f"【Gemini服务响应异常: {error_msg}】"
-
-        except requests.exceptions.RequestException as e:
-            yield f"请求失败：{e}"
-        except json.JSONDecodeError as e:
-            yield f"JSON解码错误：{e}"
-        except Exception as e:
-            yield f"发生错误：{e}"
 
     def response_with_functions(self, session_id, dialogue, functions=None):
-        logger.bind(tag=TAG).info(f"gemini暂未实现完整的工具调用（function call）")
+        yield from self._generate(dialogue, self._build_tools(functions))
-        return self.response(session_id, dialogue)
+
+    def _generate(self, dialogue, tools):
+        role_map = {"assistant": "model", "user": "user"}
+        contents: list = []
+        # 拼接对话
+        for m in dialogue:
+            r = m["role"]
+
+            if r == "assistant" and "tool_calls" in m:
+                tc = m["tool_calls"][0]
+                contents.append(
+                    {
+                        "role": "model",
+                        "parts": [
+                            {
+                                "function_call": {
+                                    "name": tc["function"]["name"],
+                                    "args": json.loads(tc["function"]["arguments"]),
+                                }
+                            }
+                        ],
+                    }
+                )
+                continue
+
+            if r == "tool":
+                contents.append(
+                    {
+                        "role": "model",
+                        "parts": [{"text": str(m.get("content", ""))}],
+                    }
+                )
+                continue
+
+            contents.append(
+                {
+                    "role": role_map.get(r, "user"),
+                    "parts": [{"text": str(m.get("content", ""))}],
+                }
+            )
+
+        stream: GenerateContentResponse = self.model.generate_content(
+            contents=contents,
+            generation_config=self.gen_cfg,
+            tools=tools,
+            stream=True,
+        )
+
+        try:
+            for chunk in stream:
+                cand = chunk.candidates[0]
+                for part in cand.content.parts:
+                    # a) 函数调用-通常是最后一段话才是函数调用
+                    if getattr(part, "function_call", None):
+                        fc = part.function_call
+                        yield None, [
+                            SimpleNamespace(
+                                id=uuid.uuid4().hex,
+                                type="function",
+                                function=SimpleNamespace(
+                                    name=fc.name,
+                                    arguments=json.dumps(
+                                        dict(fc.args), ensure_ascii=False
+                                    ),
+                                ),
+                            )
+                        ]
+                        return
+                    # b) 普通文本
+                    if getattr(part, "text", None):
+                        yield part.text if tools is None else (part.text, None)
+
+        finally:
+            if tools is not None:
+                yield None, None  # function‑mode 结束，返回哑包
+
+    # 关闭stream，预留后续打断对话功能的功能方法，官方文档推荐打断对话要关闭上一个流，可以有效减少配额计费和资源占用
+    @staticmethod
+    def _safe_finish_stream(stream: GenerateContentResponse):
+        if hasattr(stream, "resolve"):
+            stream.resolve()  # Gemini SDK version ≥ 0.5.0
+        elif hasattr(stream, "close"):
+            stream.close()  # Gemini SDK version < 0.5.0
+        else:
+            for _ in stream:  # 兜底耗尽
+                pass

main/xiaozhi-server/core/providers/llm/homeassistant/homeassistant.py

@@ -0,0 +1,71 @@
+import requests
+from requests.exceptions import RequestException
+from config.logger import setup_logging
+from core.providers.llm.base import LLMProviderBase
+
+TAG = __name__
+logger = setup_logging()
+
+
+class LLMProvider(LLMProviderBase):
+    def __init__(self, config):
+        self.agent_id = config.get("agent_id")  # 对应 agent_id
+        self.api_key = config.get("api_key")
+        self.base_url = config.get("base_url", config.get("url"))  # 默认使用 base_url
+        self.api_url = f"{self.base_url}/api/conversation/process"  # 拼接完整的 API URL
+
+    def response(self, session_id, dialogue):
+        try:
+            # home assistant语音助手自带意图，无需使用xiaozhi ai自带的，只需要把用户说的话传递给home assistant即可
+
+            # 提取最后一个 role 为 'user' 的 content
+            input_text = None
+            if isinstance(dialogue, list):  # 确保 dialogue 是一个列表
+                # 逆序遍历，找到最后一个 role 为 'user' 的消息
+                for message in reversed(dialogue):
+                    if message.get("role") == "user":  # 找到 role 为 'user' 的消息
+                        input_text = message.get("content", "")
+                        break  # 找到后立即退出循环
+
+            # 构造请求数据
+            payload = {
+                "text": input_text,
+                "agent_id": self.agent_id,
+                "conversation_id": session_id,  # 使用 session_id 作为 conversation_id
+            }
+            # 设置请求头
+            headers = {
+                "Authorization": f"Bearer {self.api_key}",
+                "Content-Type": "application/json",
+            }
+
+            # 发起 POST 请求
+            response = requests.post(self.api_url, json=payload, headers=headers)
+
+            # 检查请求是否成功
+            response.raise_for_status()
+
+            # 解析返回数据
+            data = response.json()
+            speech = (
+                data.get("response", {})
+                .get("speech", {})
+                .get("plain", {})
+                .get("speech", "")
+            )
+
+            # 返回生成的内容
+            if speech:
+                yield speech
+            else:
+                logger.bind(tag=TAG).warning("API 返回数据中没有 speech 内容")
+
+        except RequestException as e:
+            logger.bind(tag=TAG).error(f"HTTP 请求错误: {e}")
+        except Exception as e:
+            logger.bind(tag=TAG).error(f"生成响应时出错: {e}")
+
+    def response_with_functions(self, session_id, dialogue, functions=None):
+        logger.bind(tag=TAG).error(
+            f"homeassistant不支持（function call），建议使用其他意图识别"
+        )

main/xiaozhi-server/core/providers/memory/base.py

@@ -4,6 +4,7 @@ from config.logger import setup_logging
 TAG = __name__
 logger = setup_logging()
 
+
 class MemoryProviderBase(ABC):
     def __init__(self, config):
         self.config = config
@@ -20,6 +21,6 @@ class MemoryProviderBase(ABC):
         """Query memories for specific role based on similarity"""
         return "please implement query method"
 
-    def init_memory(self, role_id, llm):
+    def init_memory(self, role_id, llm, **kwargs):
-        self.role_id = role_id    
+        self.role_id = role_id
         self.llm = llm

main/xiaozhi-server/core/providers/memory/mem0ai/mem0ai.py

@@ -8,7 +8,7 @@ TAG = __name__
 
 
 class MemoryProvider(MemoryProviderBase):
-    def __init__(self, config):
+    def __init__(self, config, summary_memory=None):
         super().__init__(config)
         self.api_key = config.get("api_key", "")
         self.api_version = config.get("api_version", "v1.1")

main/xiaozhi-server/core/providers/memory/mem_local_short/mem_local_short.py

@@ -4,6 +4,7 @@ import json
 import os
 import yaml
 from config.config_loader import get_project_dir
+from config.manage_api_client import save_mem_local_short
 
 
 short_term_memory_prompt = """
@@ -72,6 +73,17 @@ short_term_memory_prompt = """
 ```
 """
 
+short_term_memory_prompt_only_content = """
+你是一个经验丰富的记忆总结者，擅长将对话内容进行总结摘要，遵循以下规则：
+1、总结user的重要信息，以便在未来的对话中提供更个性化的服务
+2、不要重复总结，不要遗忘之前记忆，除非原来的记忆超过了1800字内，否则不要遗忘、不要压缩用户的历史记忆
+3、用户操控的设备音量、播放音乐、天气、退出、不想对话等和用户本身无关的内容，这些信息不需要加入到总结中
+4、不要把设备操控的成果结果和失败结果加入到总结中，也不要把用户的一些废话加入到总结中
+5、不要为了总结而总结，如果用户的聊天没有意义，请返回原来的历史记录也是可以的
+6、只需要返回总结摘要，严格控制在1800字内
+7、不要包含代码、xml，不需要解释、注释和说明，保存记忆时仅从对话提取信息，不要混入示例内容
+"""
+
 
 def extract_json_data(json_code):
     start = json_code.find("```json")
@@ -93,17 +105,26 @@ TAG = __name__
 
 
 class MemoryProvider(MemoryProviderBase):
-    def __init__(self, config):
+    def __init__(self, config, summary_memory):
         super().__init__(config)
         self.short_momery = ""
+        self.save_to_file = True
         self.memory_path = get_project_dir() + "data/.memory.yaml"
-        self.load_memory()
+        self.load_memory(summary_memory)
 
-    def init_memory(self, role_id, llm):
+    def init_memory(
-        super().init_memory(role_id, llm)
+        self, role_id, llm, summary_memory=None, save_to_file=True, **kwargs
-        self.load_memory()
+    ):
+        super().init_memory(role_id, llm, **kwargs)
+        self.save_to_file = save_to_file
+        self.load_memory(summary_memory)
+
+    def load_memory(self, summary_memory):
+        # api获取到总结记忆后直接返回
+        if summary_memory or not self.save_to_file:
+            self.short_momery = summary_memory
+            return
 
-    def load_memory(self):
         all_memory = {}
         if os.path.exists(self.memory_path):
             with open(self.memory_path, "r", encoding="utf-8") as f:
@@ -134,7 +155,7 @@ class MemoryProvider(MemoryProviderBase):
                 msgStr += f"User: {msg.content}\n"
             elif msg.role == "assistant":
                 msgStr += f"Assistant: {msg.content}\n"
-        if len(self.short_momery) > 0:
+        if self.short_momery and len(self.short_momery) > 0:
             msgStr += "历史记忆：\n"
             msgStr += self.short_momery
 
@@ -142,16 +163,20 @@ class MemoryProvider(MemoryProviderBase):
         time_str = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
         msgStr += f"当前时间：{time_str}"
 
-        result = self.llm.response_no_stream(short_term_memory_prompt, msgStr)
+        if self.save_to_file:
-
+            result = self.llm.response_no_stream(short_term_memory_prompt, msgStr)
-        json_str = extract_json_data(result)
+            json_str = extract_json_data(result)
-        try:
+            try:
-            json_data = json.loads(json_str)  # 检查json格式是否正确
+                json.loads(json_str)  # 检查json格式是否正确
-            self.short_momery = json_str
+                self.short_momery = json_str
-        except Exception as e:
+                self.save_memory_to_file()
-            print("Error:", e)
+            except Exception as e:
-
+                print("Error:", e)
-        self.save_memory_to_file()
+        else:
+            result = self.llm.response_no_stream(
+                short_term_memory_prompt_only_content, msgStr
+            )
+            save_mem_local_short(self.role_id, result)
         logger.bind(tag=TAG).info(f"Save memory successful - Role: {self.role_id}")
 
         return self.short_momery

main/xiaozhi-server/core/providers/memory/nomem/nomem.py

@@ -1,18 +1,20 @@
-'''
+"""
 不使用记忆，可以选择此模块
-'''
+"""
+
 from ..base import MemoryProviderBase, logger
 
 TAG = __name__
 
+
 class MemoryProvider(MemoryProviderBase):
-    def __init__(self, config):
+    def __init__(self, config, summary_memory=None):
         super().__init__(config)
-      
+
     async def save_memory(self, msgs):
         logger.bind(tag=TAG).debug("nomem mode: No memory saving is performed.")
         return None
 
-    async def query_memory(self, query: str)-> str:
+    async def query_memory(self, query: str) -> str:
         logger.bind(tag=TAG).debug("nomem mode: No memory query is performed.")
-        return ""
+        return ""

main/xiaozhi-server/core/providers/tts/base.py

@@ -3,7 +3,7 @@ from config.logger import setup_logging
 import os
 from abc import ABC, abstractmethod
 from core.utils.tts import MarkdownCleaner
-from core.utils.util import audio_to_opus_data
+from core.utils.util import audio_to_data
 
 TAG = __name__
 logger = setup_logging()
@@ -53,5 +53,10 @@ class TTSProviderBase(ABC):
     async def text_to_speak(self, text, output_file):
         pass
 
+    def audio_to_pcm_data(self, audio_file_path):
+        """音频文件转换为PCM编码"""
+        return audio_to_data(audio_file_path, is_opus=False)
+
     def audio_to_opus_data(self, audio_file_path):
-        return audio_to_opus_data(audio_file_path)
+        """音频文件转换为Opus编码"""
+        return audio_to_data(audio_file_path, is_opus=True)

main/xiaozhi-server/core/providers/tts/cozecn.py

@@ -38,9 +38,13 @@ class TTSProvider(TTSProviderBase):
             "Authorization": f"Bearer {self.access_token}",
             "Content-Type": "application/json",
         }
-        response = requests.request(
+
-            "POST", self.api_url, json=request_json, headers=headers
+        try:
-        )
+            response = requests.request(
-        data = response.content
+                "POST", self.api_url, json=request_json, headers=headers
-        file_to_save = open(output_file, "wb")
+            )
-        file_to_save.write(data)
+            data = response.content
+            file_to_save = open(output_file, "wb")
+            file_to_save.write(data)
+        except Exception as e:
+            raise Exception(f"{__name__} error: {e}")

main/xiaozhi-server/core/providers/tts/custom.py

@@ -32,4 +32,6 @@ class TTSProvider(TTSProviderBase):
             with open(output_file, "wb") as file:
                 file.write(resp.content)
         else:
-            logger.bind(tag=TAG).error(f"Custom TTS请求失败: {resp.status_code} - {resp.text}")
+            error_msg = f"Custom TTS请求失败: {resp.status_code} - {resp.text}"
+            logger.bind(tag=TAG).error(error_msg)
+            raise Exception(error_msg)  # 抛出异常，让调用方捕获

main/xiaozhi-server/core/providers/tts/edge.py

@@ -20,14 +20,18 @@ class TTSProvider(TTSProviderBase):
         )
 
     async def text_to_speak(self, text, output_file):
-        communicate = edge_tts.Communicate(text, voice=self.voice)
+        try:
-        # 确保目录存在并创建空文件
+            communicate = edge_tts.Communicate(text, voice=self.voice)
-        os.makedirs(os.path.dirname(output_file), exist_ok=True)
+            # 确保目录存在并创建空文件
-        with open(output_file, "wb") as f:
+            os.makedirs(os.path.dirname(output_file), exist_ok=True)
-            pass
+            with open(output_file, "wb") as f:
+                pass
 
-        # 流式写入音频数据
+            # 流式写入音频数据
-        with open(output_file, "ab") as f:  # 改为追加模式避免覆盖
+            with open(output_file, "ab") as f:  # 改为追加模式避免覆盖
-            async for chunk in communicate.stream():
+                async for chunk in communicate.stream():
-                if chunk["type"] == "audio":  # 只处理音频数据块
+                    if chunk["type"] == "audio":  # 只处理音频数据块
-                    f.write(chunk["data"])
+                        f.write(chunk["data"])
+        except Exception as e:
+            error_msg = f"Edge TTS请求失败: {e}"
+            raise Exception(error_msg)  # 抛出异常，让调用方捕获

main/xiaozhi-server/core/providers/tts/fishspeech.py

@@ -85,16 +85,22 @@ class TTSProvider(TTSProviderBase):
     def __init__(self, config, delete_audio_file):
         super().__init__(config, delete_audio_file)
 
-        self.reference_id = config.get("reference_id")
+        self.reference_id = (
+            None if not config.get("reference_id") else config.get("reference_id")
+        )
         self.reference_audio = parse_string_to_list(config.get("reference_audio"))
         self.reference_text = parse_string_to_list(config.get("reference_text"))
-        self.format = config.get("format", "wav")
+        self.format = config.get("response_format", "wav")
 
         self.api_key = config.get("api_key", "YOUR_API_KEY")
         have_key = check_model_key("FishSpeech TTS", self.api_key)
         if not have_key:
             return
-        self.normalize = config.get("normalize", True)
+        self.normalize = str(config.get("normalize", True)).lower() in (
+            "true",
+            "1",
+            "yes",
+        )
 
         # 处理空字符串的情况
         channels = config.get("channels", "1")
@@ -177,5 +183,7 @@ class TTSProvider(TTSProviderBase):
                 audio_file.write(audio_content)
 
         else:
-            print(f"Request failed with status code {response.status_code}")
+            error_msg = f"Request failed with status code {response.status_code}"
+            print(error_msg)
             print(response.json())
+            raise Exception(error_msg)

main/xiaozhi-server/core/providers/tts/gpt_sovits_v2.py

@@ -105,6 +105,6 @@ class TTSProvider(TTSProviderBase):
             with open(output_file, "wb") as file:
                 file.write(resp.content)
         else:
-            logger.bind(tag=TAG).error(
+            error_msg = f"GPT_SoVITS_V2 TTS请求失败: {resp.status_code} - {resp.text}"
-                f"GPT_SoVITS_V2 TTS请求失败: {resp.status_code} - {resp.text}"
+            logger.bind(tag=TAG).error(error_msg)
-            )
+            raise Exception(error_msg)

main/xiaozhi-server/core/providers/tts/gpt_sovits_v3.py

@@ -64,6 +64,7 @@ class TTSProvider(TTSProviderBase):
             with open(output_file, "wb") as file:
                 file.write(resp.content)
         else:
-            logger.bind(tag=TAG).error(
+            error_msg = f"GPT_SoVITS_V3 TTS请求失败: {resp.status_code} - {resp.text}"
-                f"GPT_SoVITS_V3 TTS请求失败: {resp.status_code} - {resp.text}"
+            logger.bind(tag=TAG).error(error_msg)
-            )
+            raise Exception(error_msg)
+

main/xiaozhi-server/core/providers/tts/siliconflow.py

@@ -39,9 +39,12 @@ class TTSProvider(TTSProviderBase):
             "Authorization": f"Bearer {self.access_token}",
             "Content-Type": "application/json",
         }
-        response = requests.request(
+        try:
-            "POST", self.api_url, json=request_json, headers=headers
+            response = requests.request(
-        )
+                "POST", self.api_url, json=request_json, headers=headers
-        data = response.content
+            )
-        file_to_save = open(output_file, "wb")
+            data = response.content
-        file_to_save.write(data)
+            file_to_save = open(output_file, "wb")
+            file_to_save.write(data)
+        except Exception as e:
+            raise Exception(f"{__name__} error: {e}")

main/xiaozhi-server/core/providers/tts/ttson.py

@@ -58,8 +58,8 @@ class TTSProvider(TTSProviderBase):
 
         resp = requests.request("POST", url, data=payload)
         if resp.status_code != 200:
-            logger.bind(tag=TAG).error(f"TTS请求失败: {resp.text}")
+            logger.bind(tag=TAG).error(f"TTSON 请求失败: {resp.text}")
-            return None
+            raise Exception(f"{__name__}: TTS请求失败")
         resp_json = resp.json()
         try:
             result = (
@@ -71,13 +71,15 @@ class TTSProvider(TTSProviderBase):
                 + "&voice_audio_path="
                 + resp_json["voice_path"]
             )
+
+            audio_content = requests.get(result)
+            with open(output_file, "wb") as f:
+                f.write(audio_content.content)
+                return True
+            voice_path = resp_json.get("voice_path")
+            des_path = output_file
+            shutil.move(voice_path, des_path)
+
         except Exception as e:
             print("error:", e)
-
+            raise Exception(f"{__name__}: TTS请求失败")
-        audio_content = requests.get(result)
-        with open(output_file, "wb") as f:
-            f.write(audio_content.content)
-            return True
-        voice_path = resp_json.get("voice_path")
-        des_path = output_file
-        shutil.move(voice_path, des_path)

main/xiaozhi-server/core/utils/dialogue.py

@@ -33,7 +33,13 @@ class Dialogue:
             dialogue.append({"role": m.role, "tool_calls": m.tool_calls})
         elif m.role == "tool":
             dialogue.append(
-                {"role": m.role, "tool_call_id": m.tool_call_id, "content": m.content}
+                {
+                    "role": m.role,
+                    "tool_call_id": (
+                        str(uuid.uuid4()) if m.tool_call_id is None else m.tool_call_id
+                    ),
+                    "content": m.content,
+                }
             )
         else:
             dialogue.append({"role": m.role, "content": m.content})
@@ -69,7 +75,8 @@ class Dialogue:
 
         if system_message:
             enhanced_system_prompt = (
-                f"{system_message.content}\n\n" f"相关记忆：\n{memory_str}"
+                f"{system_message.content}\n\n"
+                f"以下是用户的历史记忆：\n```\n{memory_str}\n```"
             )
             dialogue.append({"role": "system", "content": enhanced_system_prompt})
 

main/xiaozhi-server/core/utils/util.py

@@ -319,6 +319,7 @@ def initialize_modules(
         modules["memory"] = memory.create_instance(
             memory_type,
             config["Memory"][select_memory_module],
+            config.get('summaryMemory', None),
         )
         logger.bind(tag=TAG).info(f"初始化组件: memory成功 {select_memory_module}")
 
@@ -862,8 +863,7 @@ def analyze_emotion(text):
     return top_emotions[0]  # 如果都不在优先级列表里，返回第一个
 
 
-def audio_to_opus_data(audio_file_path):
+def audio_to_data(audio_file_path, is_opus=True):
-    """音频文件转换为Opus编码"""
     # 获取文件后缀名
     file_type = os.path.splitext(audio_file_path)[1]
     if file_type:
@@ -889,7 +889,7 @@ def audio_to_opus_data(audio_file_path):
     frame_duration = 60  # 60ms per frame
     frame_size = int(16000 * frame_duration / 1000)  # 960 samples/frame
 
-    opus_datas = []
+    datas = []
     # 按帧处理所有音频数据（包括最后一帧可能补零）
     for i in range(0, len(raw_data), frame_size * 2):  # 16bit=2bytes/sample
         # 获取当前帧的二进制数据
@@ -899,18 +899,24 @@ def audio_to_opus_data(audio_file_path):
         if len(chunk) < frame_size * 2:
             chunk += b"\x00" * (frame_size * 2 - len(chunk))
 
-        # 转换为numpy数组处理
+        if is_opus:
-        np_frame = np.frombuffer(chunk, dtype=np.int16)
+            # 转换为numpy数组处理
+            np_frame = np.frombuffer(chunk, dtype=np.int16)
+            # 编码Opus数据
+            frame_data = encoder.encode(np_frame.tobytes(), frame_size)
+        else:
+            frame_data = chunk if isinstance(chunk, bytes) else bytes(chunk)
 
-        # 编码Opus数据
+        datas.append(frame_data)
-        opus_data = encoder.encode(np_frame.tobytes(), frame_size)
-        opus_datas.append(opus_data)
 
-    return opus_datas, duration
+    return datas, duration
 
 
 def check_vad_update(before_config, new_config):
-    if new_config["selected_module"].get("VAD") is None:
+    if (
+        new_config.get("selected_module") is None
+        or new_config["selected_module"].get("VAD") is None
+    ):
         return False
     update_vad = False
     current_vad_module = before_config["selected_module"]["VAD"]
@@ -925,13 +931,15 @@ def check_vad_update(before_config, new_config):
         if "type" not in new_config["VAD"][new_vad_module]
         else new_config["VAD"][new_vad_module]["type"]
     )
-    print(f"前vad:{current_vad_type}，后vad:{new_vad_type}")
     update_vad = current_vad_type != new_vad_type
     return update_vad
 
 
 def check_asr_update(before_config, new_config):
-    if new_config["selected_module"].get("ASR") is None:
+    if (
+        new_config.get("selected_module") is None
+        or new_config["selected_module"].get("ASR") is None
+    ):
         return False
     update_asr = False
     current_asr_module = before_config["selected_module"]["ASR"]
@@ -946,6 +954,5 @@ def check_asr_update(before_config, new_config):
         if "type" not in new_config["ASR"][new_asr_module]
         else new_config["ASR"][new_asr_module]["type"]
     )
-    print(f"前asr:{current_asr_type}，后asr:{new_asr_type}")
     update_asr = current_asr_type != new_asr_type
     return update_asr