Compare commits

..
11 Commits
Author SHA1 Message Date
HuangandGitHub c796af7d07 Hot fix (#415)
* fix:找不到function卡住bug

* fix:客户端拾音模式日志bug
2025-03-18 18:42:17 +08:00
bd556c33a0 fix:找不到function卡住bug (#408)
Co-authored-by: hrz <1710360675@qq.com>
2025-03-18 14:26:27 +08:00
878809ecb4 update:修复iot的bug (#407)
* fix: 修复manual模式无法识别 (#404)

* feat(docs): 新增Issues模板

* fix: 补全core依赖

* fix: 修复manual模式无法识别

* 去除重复依赖.txt

已经有torch和torchaudio

---------

Co-authored-by: 欣南科技 <huangrongzhuang@xin-nan.com>

* 修复iot功能中表达式问题 (#400)

* Custom paths asr tts (#388)

* #164 自定义asr、tts缓存目录,项目启动自动创建目录

* #164 自定义asr、tts缓存目录,项目启动自动创建目录

* fix:修复语音无法找到新配置项output_file的bug

* fix:电脑不支持iot音量控制bug

---------

Co-authored-by: Junsen <66542771+Huang-junsen@users.noreply.github.com>
Co-authored-by: tang <tangyiyong@gmail.com>
Co-authored-by: shudongW <178200623@qq.com>
Co-authored-by: hrz <1710360675@qq.com>
2025-03-18 13:25:34 +08:00
af2ae11eb1 update:更新视频demo (#397)
Co-authored-by: hrz <1710360675@qq.com>
2025-03-18 01:35:18 +08:00
5aa4199dfb add:支持IOT功能
* 自动注册iot消息为function call函数,并通过function call控制设备

* 增加配置项,控制iot自动注册function call功能的开启

* 增加function call action的 Error code

* update:补充音量太大声、太小声的意图动作

* update:支持IOT功能

---------

Co-authored-by: 玄凤科技 <eric230308@gmail.com>
Co-authored-by: hrz <1710360675@qq.com>
2025-03-18 00:34:16 +08:00
yanyigeandGitHub 773790c8ec docs: 修复致开发者的公开信文档中的小问题 (#392) 2025-03-18 00:20:27 +08:00
8a066163fe update:和风天气查询 (#387)
* Update get_weather.py (#381)

更新天气插件,增加:通过当前IP获取用户实时位置的功能

* 重构天气查询插件 (#382)

* feat: 添加获取ip信息工具函数

* refactor: 重构天气查询插件

使用和风天气作为数据源
功能更完善

---------

Co-authored-by: 欣南科技 <huangrongzhuang@xin-nan.com>

* update:天气查询优化

---------

Co-authored-by: xiaowu911 <51016996+xiaowu911@users.noreply.github.com>
Co-authored-by: Jad <i@nocilol.me>
Co-authored-by: hrz <1710360675@qq.com>
2025-03-17 14:20:40 +08:00
JunsenandGitHub 6eac25425c feat(docs): 新增Issues模板 (#383) 2025-03-17 09:45:50 +08:00
c0f1981d6f Server music plugins (#380)
* update:intent_llm音乐插件化调用

* update:播放音乐插件改造

* update:完成音乐播放插件化改造

---------

Co-authored-by: hrz <1710360675@qq.com>
2025-03-17 02:13:10 +08:00
8376b6f0f2 更新GPT_SOVITS_V3接口 (#375)
* ollama去除回复中的think字段 (#373)

* ollama调用修复,并且去除回复中的think字段(deepseek)

* ollama调用修复,并且去除回复中的think字段(deepseek)

* ollama去除回复中的think字段(deepseek)
deepseek 14b,qwen2.5测试通过

* ollama去除回复中的think字段(deepseek)
deepseek 14b,qwen2.5测试通过

* Update ollama.py

---------

Co-authored-by: aileenfun <105077614@qq.com>
Co-authored-by: 欣南科技 <huangrongzhuang@xin-nan.com>

* fix:GPT_SOVITS_V3-20250228 API (#312)

---------

Co-authored-by: aileenfun <alanliu0618@gmail.com>
Co-authored-by: aileenfun <105077614@qq.com>
Co-authored-by: Miyo <101488652+Miyokiss@users.noreply.github.com>
2025-03-16 16:04:27 +08:00
490a66d3fe Add deploy pic (#354)
* update:添加部署方案图

* update:部署参考图

---------

Co-authored-by: hrz <1710360675@qq.com>
2025-03-15 16:17:25 +08:00
34 changed files with 1006 additions and 405 deletions
+34
View File
@@ -0,0 +1,34 @@
---
name: Bug 报告(Bug Report
about: 反馈项目中的缺陷或问题
title: "[Bug] 简短描述问题"
labels: bug
assignees: ''
---
## 🐛 问题描述
<!-- 清晰简洁地描述问题是什么 -->
## 🔍 复现步骤
<!-- 详细描述复现问题的步骤 -->
1. 打开 '...'
2. 点击 '...'
3. 滚动到 '...'
4. 看到错误
## 🤔 预期行为
<!-- 简要描述预期的正确行为 -->
## 😯 截图
<!-- 如果适用,添加问题的截图 -->
## 🖥️ 环境信息
- 操作系统: [例如 Windows 10]
- 浏览器: [例如 Chrome 89]
- 项目版本: [例如 1.0.0]
- Python版本: [例如 3.9.13]
- Jdk版本:[例如 java 21.0.5 2024-10-15 LTS]
- Nodejs版本:[例如 v20.14.0]
## 📋 其他信息
<!-- 在此添加关于此问题的任何其他上下文信息 -->
@@ -0,0 +1,19 @@
---
name: 代码优化建议(Code Improvement
about: 提出对现有代码的优化或改进建议
title: "[Improvement] 简短描述改进内容"
labels: refactor
assignees: ''
---
## 💡 改进描述
<!-- 描述需要改进的代码部分以及存在的问题 -->
## 🌟 改进建议
<!-- 提出具体的改进方案或思路 -->
## 🛠️ 相关代码
<!-- 如果可能,提供相关代码片段或链接 -->
## 📋 其他信息
<!-- 在此添加任何其他相关信息 -->
@@ -0,0 +1,16 @@
---
name: 文档改进建议(Documentation Improvement
about: 提出对项目文档的改进或补充建议
title: "[Docs] 简短描述改进内容"
labels: documentation
assignees: ''
---
## 📚 改进描述
<!-- 描述需要改进的文档部分以及存在的问题 -->
## ✨ 改进建议
<!-- 提出具体的改进方案或内容 -->
## 📋 其他信息
<!-- 在此添加任何其他相关信息 -->
+19
View File
@@ -0,0 +1,19 @@
---
name: 功能请求(Feature Request
about: 提出新的功能或改进建议
title: "[Feature] 简短描述功能"
labels: enhancement
assignees: ''
---
## 🚀 需求描述
<!-- 清晰简洁地描述你希望发生什么 -->
## 🎯 解决方案
<!-- 描述你认为可行的解决方案或实现方式 -->
## 📝 备选方案
<!-- 描述你考虑过的替代方案或功能 -->
## 📋 其他信息
<!-- 在此添加关于需求的任何其他上下文信息 -->
+25 -8
View File
@@ -57,6 +57,13 @@
</picture>
</a>
</td>
<td>
<a href="https://www.bilibili.com/video/BV1kgA2eYEQ9" target="_blank">
<picture>
<img alt="成本最低配置" src="docs/images/demo4.png" />
</picture>
</a>
</td>
</tr>
<tr>
<td>
@@ -74,9 +81,16 @@
</a>
</td>
<td>
<a href="https://www.bilibili.com/video/BV1kgA2eYEQ9" target="_blank">
<a href="https://www.bilibili.com/video/BV1Z8XuYZEAS" target="_blank">
<picture>
<img alt="成本最低配置" src="docs/images/demo4.png" />
<img alt="天气插件" src="docs/images/demo8.png" />
</picture>
</a>
</td>
<td>
<a href="https://www.bilibili.com/video/BV178XuYfEpi" target="_blank">
<picture>
<img alt="IOT指令控制设备" src="docs/images/demo9.png" />
</picture>
</a>
</td>
@@ -92,9 +106,11 @@
- **硬件**:一套兼容 `xiaozhi-esp32`
的硬件设备(具体型号请参考 [此处](https://rcnv1t9vps13.feishu.cn/wiki/DdgIw4BUgivWDPkhMj1cGIYCnRf))。
- **电脑或服务器**至少 4 核 CPU、8G 内存的电脑。
- **电脑或服务器**建议 4 核 CPU、8G 内存的电脑。如果开启ASR也使用API,可运行在2核CPU、2G内存的服务器中。
- **固件编译**:请将本后端服务的接口地址更新至 `xiaozhi-esp32` 项目中,再重新编译`xiaozhi-esp32`固件并烧录到设备上。
![图片](docs/images/deploy.png)
如果你没有esp32相关的硬件设备,但是非常想体验该项目,可以使用以下的项目让你的电脑、手机模拟成esp32设备。
- [小智安卓端](https://github.com/TOM88812/xiaozhi-android-client)
@@ -141,12 +157,13 @@ server:
支持 EdgeTTS(默认)、火山引擎豆包 TTS 等多种 TTS 接口,满足语音合成需求。
- **记忆功能**
支持超长记忆、本地总结记忆、无记忆三种模式,满足不同场景需求。
- **IOT功能**
支持管理注册设备IOT功能,支持基于对话上下文语境下的智能物联网控制。
### 正在开发 🚧
- 多种心情模式
- 智控台webui
- iot功能
想了解具体开发进度,[请点击这里](https://github.com/users/xinnan-tech/projects/3)
@@ -212,10 +229,10 @@ server:
### Memory 记忆存储
| 类型 | 平台名称 | 使用方式 | 收费模式 | 备注 |
|:------:|:---------------:|:----:|:--------:|:--:|
| Memory | mem0ai | 接口调用 | 100次/月额度 | |
| Memory | mem_local_short | 本地总结 | 免费 | |
| 类型 | 平台名称 | 使用方式 | 收费模式 | 备注 |
|:------:|:---------------:|:----:|:---------:|:--:|
| Memory | mem0ai | 接口调用 | 1000次/月额度 | |
| Memory | mem_local_short | 本地总结 | 免费 | |
---
+2
View File
@@ -1,3 +1,5 @@
# 部署方案参考
![图片](images/deploy.png)
# 方式一:docker快速部署
docker镜像已支持x86架构、arm64架构的CPU,支持在国产操作系统上运行。
+1 -1
View File
@@ -35,7 +35,7 @@ John2025.3.11,广州
### 1、成为普通贡献者
Fork 项目,提交 PR,由开发者审核后合入主分支。
Fork 项目,提交 PR,由开发者审核后合入主分支。
### 2、成为开发者
Binary file not shown.

After

Width:  |  Height:  |  Size: 280 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 279 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 576 KiB

+1 -1
View File
@@ -18,6 +18,6 @@ xiaozhi-esp32-server
# manager-web 、manager-api接口协议
[manager前后端接口协议](https://app.apifox.com/invite/project?token=H_8qhgfjUeaAL0wybghgU)
[manager前后端接口协议](https://app.apifox.com/invite/project?token=eXg2_tUv85q-gc3ZRowmn)
[前端页面设计图](https://codesign.qq.com/app/s/526108506410828)
+40 -31
View File
@@ -35,10 +35,7 @@ log:
log_file: "server.log"
# 设置数据文件路径
data_dir: data
iot:
Speaker:
# 设置esp32的音量,范围0-100
volume: 80
xiaozhi:
type: hello
version: 1
@@ -102,6 +99,14 @@ Intent:
- change_role
- get_weather
# 插件的基础配置
plugins:
# 获取天气插件的配置,这里填写你的api_key
# 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" }
Memory:
mem0ai:
type: mem0ai
@@ -109,10 +114,10 @@ Memory:
# 每月有1000次免费调用
api_key: 你的mem0ai api key
nomem:
# 不想使用记忆功能,可以使用nomem
# 不想使用记忆功能,可以使用nomem
type: nomem
mem_local_short:
# 本地记忆功能,通过selected_module的llm总结,数据保存在本地,不会上传到服务器
# 本地记忆功能,通过selected_module的llm总结,数据保存在本地,不会上传到服务器
type: mem_local_short
ASR:
@@ -220,7 +225,7 @@ TTS:
# 定义TTS API类型
type: edge
voice: zh-CN-XiaoxiaoNeural
output_file: tmp/
output_dir: tmp/
DoubaoTTS:
# 定义TTS API类型
type: doubao
@@ -230,7 +235,7 @@ TTS:
# 地址:https://console.volcengine.com/speech/service/8
api_url: https://openspeech.bytedance.com/api/v1/tts
voice: BV001_streaming
output_file: tmp/
output_dir: tmp/
authorization: "Bearer;"
appid: 你的火山引擎语音合成服务appid
access_token: 你的火山引擎语音合成服务access_token
@@ -241,7 +246,7 @@ TTS:
# token申请地址 https://cloud.siliconflow.cn/account/ak
model: FunAudioLLM/CosyVoice2-0.5B
voice: FunAudioLLM/CosyVoice2-0.5B:alex
output_file: tmp/
output_dir: tmp/
access_token: 你的硅基流动API密钥
response_format: wav
CozeCnTTS:
@@ -249,7 +254,7 @@ TTS:
# COZECN TTS
# token申请地址 https://www.coze.cn/open/oauth/pats
voice: 7426720361733046281
output_file: tmp/
output_dir: tmp/
access_token: 你的coze web key
response_format: wav
FishSpeech:
@@ -262,7 +267,7 @@ TTS:
#--decoder-config-name firefly_gan_vq
#--compile
type: fishspeech
output_file: tmp/
output_dir: tmp/
response_format: wav
reference_id: null
reference_audio: ["/tmp/test.wav",]
@@ -286,7 +291,7 @@ TTS:
#python api_v2.py -a 127.0.0.1 -p 9880 -c GPT_SoVITS/configs/caixukun.yaml
type: gpt_sovits_v2
url: "http://127.0.0.1:9880/tts"
output_file: tmp/
output_dir: tmp/
text_lang: "auto"
ref_audio_path: "caixukun.wav"
prompt_text: ""
@@ -306,20 +311,24 @@ TTS:
repetition_penalty: 1.35
aux_ref_audio_paths: []
GPT_SOVITS_V3:
# 定义TTS API类型 GPT-SoVITS-v3lora-20250228
#启动tts方法:
#python api.py
type: gpt_sovits_v3
url: "http://127.0.0.1:9880/tts"
output_file: tmp/
text_lang: "auto"
ref_audio_path: "caixukun.wav"
prompt_lang: "zh"
url: "http://127.0.0.1:9880"
output_dir: tmp/
text_language: "auto"
refer_wav_path: "caixukun.wav"
prompt_language: "zh"
prompt_text: ""
top_k: 5
top_p: 1
temperature: 1
sample_steps: 16
media_type: "wav"
streaming_mode: false
threshold: 30
top_k: 15
top_p: 1.0
temperature: 1.0
cut_punc: ""
speed: 1.0
inp_refs: []
sample_steps: 32
if_sr: false
MinimaxTTS:
# Minimax语音合成服务,需要先在minimax平台创建账户充值,并获取登录信息
# 平台地址:https://platform.minimaxi.com/
@@ -328,7 +337,7 @@ TTS:
# api_key地址:https://platform.minimaxi.com/user-center/basic-information/interface-key
# 定义TTS API类型
type: minimax
output_file: tmp/
output_dir: tmp/
group_id: 你的minimax平台groupID
api_key: 你的minimax平台接口密钥
model: "speech-01-turbo"
@@ -365,7 +374,7 @@ TTS:
# token地址:https://nls-portal.console.aliyun.com/overview
# 定义TTS API类型
type: aliyun
output_file: tmp/
output_dir: tmp/
appkey: 你的阿里云智能语音交互服务项目Appkey
token: 你的阿里云智能语音交互服务AccessToken,临时的24小时,要长期用下方的access_key_idaccess_key_secret
voice: xiaoyun
@@ -388,7 +397,7 @@ TTS:
api_url: https://api.302ai.cn/doubao/tts_hd
authorization: "Bearer "
voice: "zh_female_wanwanxiaohe_moon_bigtts"
output_file: tmp/
output_dir: tmp/
access_token: "你的302API密钥"
ACGNTTS:
#在线网址:https://acgn.ttson.cn/
@@ -405,7 +414,7 @@ TTS:
to_lang: ZH
url: https://u95167-bd74-2aef8085.westx.seetacloud.com:8443/flashsummary/tts?token=
format: mp3
output_file: tmp/
output_dir: tmp/
emotion: 1
OpenAITTS:
# openai官方文本转语音服务,可支持全球大多数语种
@@ -419,7 +428,7 @@ TTS:
voice: onyx
# 语速范围0.25-4.0
speed: 1
output_file: tmp/
output_dir: tmp/
CustomTTS:
# 自定义的TTS接口服务,请求参数可自定义
# 要求接口使用GET方式请求,并返回音频文件
@@ -434,7 +443,7 @@ TTS:
headers: # 自定义请求头
# Authorization: Bearer xxxx
format: wav # 接口返回的音频格式
output_file: tmp/
output_dir: tmp/
# 模块测试配置
module_test:
test_sentences: # 自定义测试语句
@@ -457,4 +466,4 @@ manager:
enabled: false
ip: 0.0.0.0
port: 8002
use_private_config: false
use_private_config: false
+46 -3
View File
@@ -7,9 +7,48 @@ from core.utils.util import read_config, get_project_dir
default_config_file = "config.yaml"
def ensure_directories(config):
"""确保所有配置路径存在"""
dirs_to_create = set()
project_dir = get_project_dir() # 获取项目根目录
# 日志文件目录
log_dir = config.get('log', {}).get('log_dir', 'tmp')
dirs_to_create.add(os.path.join(project_dir, log_dir))
# ASR/TTS模块输出目录
for module in ['ASR', 'TTS']:
for provider in config.get(module, {}).values():
output_dir = provider.get('output_dir', '')
if output_dir:
dirs_to_create.add(output_dir)
# 根据selected_module创建模型目录
selected_modules = config.get('selected_module', {})
for module_type in ['ASR', 'LLM', 'TTS']:
selected_provider = selected_modules.get(module_type)
if not selected_provider:
continue
provider_config = config.get(module_type, {}).get(selected_provider, {})
output_dir = provider_config.get('output_dir')
if output_dir:
full_model_dir = os.path.join(project_dir, output_dir)
dirs_to_create.add(full_model_dir)
# 统一创建目录(保留原data目录创建)
for dir_path in dirs_to_create:
try:
os.makedirs(dir_path, exist_ok=True)
except PermissionError:
print(f"警告:无法创建目录 {dir_path},请检查写入权限")
def get_config_file():
global default_config_file
# 判断是否存在私有配置文件
"""获取配置文件路径,优先使用私有配置文件(若存在)。
Returns:
str: 配置文件路径(相对路径或默认路径)
"""
config_file = default_config_file
if os.path.exists(get_project_dir() + "data/." + default_config_file):
config_file = "data/." + default_config_file
@@ -20,9 +59,13 @@ def load_config():
"""加载配置文件"""
parser = argparse.ArgumentParser(description="Server configuration")
config_file = get_config_file()
parser.add_argument("--config_path", type=str, default=config_file)
args = parser.parse_args()
return read_config(args.config_path)
config = read_config(args.config_path)
# 初始化目录
ensure_directories(config)
return config
def update_config(config):
@@ -67,7 +110,7 @@ def find_missing_keys(new_config, old_config, parent_key=''):
def check_config_file():
old_config_file = get_config_file()
global default_config_file
if not old_config_file.startswith('data'):
if not 'data' in old_config_file:
return
old_config = read_config(get_project_dir() + old_config_file)
new_config = read_config(get_project_dir() + default_config_file)
+49 -32
View File
@@ -5,13 +5,15 @@ import time
import queue
import asyncio
import traceback
from config.logger import setup_logging
import threading
import websockets
from typing import Dict, Any
import plugins_func.loadplugins
from config.logger import setup_logging
from core.utils.dialogue import Message, Dialogue
from core.handle.textHandle import handleTextMessage
from core.utils.util import get_string_no_punctuation_or_emoji, extract_json_from_string
from core.utils.util import get_string_no_punctuation_or_emoji, extract_json_from_string, get_ip_info
from concurrent.futures import ThreadPoolExecutor, TimeoutError
from core.handle.sendAudioHandle import sendAudioMessage
from core.handle.receiveAudioHandle import handleAudioMessage
@@ -20,7 +22,6 @@ from plugins_func.register import Action
from config.private_config import PrivateConfig
from core.auth import AuthMiddleware, AuthenticationError
from core.utils.auth_code_gen import AuthCodeGenerator
import plugins_func.loadplugins
TAG = __name__
@@ -30,13 +31,15 @@ class TTSException(RuntimeError):
class ConnectionHandler:
def __init__(self, config: Dict[str, Any], _vad, _asr, _llm, _tts, _music, _memory, _intent):
def __init__(self, config: Dict[str, Any], _vad, _asr, _llm, _tts, _memory, _intent):
self.config = config
self.logger = setup_logging()
self.auth = AuthMiddleware(config)
self.websocket = None
self.headers = None
self.client_ip = None
self.client_ip_info = {}
self.session_id = None
self.prompt = None
self.welcome_msg = None
@@ -91,21 +94,18 @@ class ConnectionHandler:
self.private_config = None
self.auth_code_gen = AuthCodeGenerator.get_instance()
self.is_device_verified = False # 添加设备验证状态标志
self.music_handler = _music
self.close_after_chat = False # 是否在聊天结束后关闭连接
self.use_function_call_mode = False
if self.config["selected_module"]["Intent"] == 'function_call':
self.use_function_call_mode = True
self.func_handler = FunctionHandler(self.config)
async def handle_connection(self, ws):
try:
# 获取并验证headers
self.headers = dict(ws.request.headers)
# 获取客户端ip地址
client_ip = ws.remote_address[0]
self.logger.bind(tag=TAG).info(f"{client_ip} conn - Headers: {self.headers}")
self.client_ip = ws.remote_address[0]
self.logger.bind(tag=TAG).info(f"{self.client_ip} conn - Headers: {self.headers}")
# 进行认证
await self.auth.authenticate(self.headers)
@@ -149,6 +149,7 @@ class ConnectionHandler:
self.welcome_msg["session_id"] = self.session_id
await self.websocket.send(json.dumps(self.welcome_msg))
# 异步初始化
await self.loop.run_in_executor(None, self._initialize_components)
# tts 消化线程
@@ -189,8 +190,14 @@ class ConnectionHandler:
self.prompt = self.config["prompt"]
if self.private_config:
self.prompt = self.private_config.private_config.get("prompt", self.prompt)
self.client_ip_info = get_ip_info(self.client_ip)
self.logger.bind(tag=TAG).info(f"Client ip info: {self.client_ip_info}")
self.prompt = self.prompt + f"\n我在:{self.client_ip_info}"
self.dialogue.put(Message(role="system", content=self.prompt))
self.func_handler = FunctionHandler(self.config)
def change_system_prompt(self, prompt):
self.prompt = prompt
# 找到原来的role==system,替换原来的系统提示
@@ -296,7 +303,7 @@ class ConnectionHandler:
self.logger.bind(tag=TAG).debug(json.dumps(self.dialogue.get_llm_dialogue(), indent=4, ensure_ascii=False))
return True
def chat_with_function_calling(self, query, tool_call = False):
def chat_with_function_calling(self, query, tool_call=False):
self.logger.bind(tag=TAG).debug(f"Chat with function calling start: {query}")
"""Chat with function calling for intent detection using streaming"""
if self.isNeedAuth():
@@ -304,7 +311,7 @@ class ConnectionHandler:
future = asyncio.run_coroutine_threadsafe(self._check_and_broadcast_auth_code(), self.loop)
future.result()
return True
if not tool_call:
self.dialogue.put(Message(role="user", content=query))
@@ -313,7 +320,7 @@ class ConnectionHandler:
response_message = []
processed_chars = 0 # 跟踪已处理的字符位置
try:
start_time = time.time()
@@ -321,7 +328,7 @@ class ConnectionHandler:
future = asyncio.run_coroutine_threadsafe(self.memory.query_memory(query), self.loop)
memory_str = future.result()
#self.logger.bind(tag=TAG).info(f"对话记录: {self.dialogue.get_llm_dialogue_with_memory(memory_str)}")
# self.logger.bind(tag=TAG).info(f"对话记录: {self.dialogue.get_llm_dialogue_with_memory(memory_str)}")
# 使用支持functions的streaming接口
llm_responses = self.llm.response_with_functions(
@@ -344,8 +351,8 @@ class ConnectionHandler:
content_arguments = ""
for response in llm_responses:
content, tools_call = response
if content is not None and len(content)>0:
if len(response_message)<=0 and (content=="```" or "<tool_call>" in content):
if content is not None and len(content) > 0:
if len(response_message) <= 0 and (content == "```" or "<tool_call>" in content):
tool_call_flag = True
if tools_call is not None:
@@ -359,7 +366,7 @@ class ConnectionHandler:
if content is not None and len(content) > 0:
if tool_call_flag:
content_arguments+=content
content_arguments += content
else:
response_message.append(content)
@@ -415,16 +422,17 @@ class ConnectionHandler:
else:
function_arguments = json.loads(function_arguments)
if not bHasError:
self.logger.bind(tag=TAG).info(f"function_name={function_name}, function_id={function_id}, function_arguments={function_arguments}")
self.logger.bind(tag=TAG).info(
f"function_name={function_name}, function_id={function_id}, function_arguments={function_arguments}")
function_call_data = {
"name": function_name,
"id": function_id,
"arguments": function_arguments
}
result = self.func_handler.handle_llm_function_call(self, function_call_data)
self._handle_function_result(result, function_call_data, text_index+1)
self._handle_function_result(result, function_call_data, text_index + 1)
# 处理最后剩余的文本
# 处理最后剩余的文本
full_text = "".join(response_message)
remaining_text = full_text[processed_chars:]
if remaining_text:
@@ -436,7 +444,7 @@ class ConnectionHandler:
self.tts_queue.put(future)
# 存储对话内容
if len(response_message)>0:
if len(response_message) > 0:
self.dialogue.put(Message(role="assistant", content="".join(response_message)))
self.llm_finish_task = True
@@ -445,31 +453,40 @@ class ConnectionHandler:
return True
def _handle_function_result(self, result, function_call_data, text_index):
if result.action == Action.RESPONSE: # 直接回复前端
if result.action == Action.RESPONSE: # 直接回复前端
text = result.response
self.recode_first_last_text(text, text_index)
future = self.executor.submit(self.speak_and_play, text, text_index)
self.tts_queue.put(future)
self.dialogue.put(Message(role="assistant", content=text))
if result.action == Action.REQLLM: # 调用函数后再请求llm生成回复
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 and len(text) > 0:
function_id = function_call_data["id"]
function_name = function_call_data["name"]
function_arguments = function_call_data["arguments"]
self.dialogue.put(Message(role='assistant',
tool_calls=[{"id": function_id,
"function": {"arguments": function_arguments,"name": function_name},
"type": 'function',
"index": 0}]))
tool_calls=[{"id": function_id,
"function": {"arguments": function_arguments,
"name": function_name},
"type": 'function',
"index": 0}]))
self.dialogue.put(Message(role="tool", tool_call_id=function_id, content=text))
self.chat_with_function_calling(text, tool_call=True)
if result.action == Action.NOTFOUND:
text = result.response
elif result.action == Action.NOTFOUND:
text = result.result
self.recode_first_last_text(text, text_index)
future = self.executor.submit(self.speak_and_play, text, text_index)
self.tts_queue.put(future)
self.dialogue.put(Message(role="assistant", content=text))
else:
text = result.result
self.recode_first_last_text(text, text_index)
future = self.executor.submit(self.speak_and_play, text, text_index)
self.tts_queue.put(future)
self.dialogue.put(Message(role="assistant", content=text))
def _tts_priority_thread(self):
while not self.stop_event.is_set():
@@ -50,6 +50,7 @@ class FunctionHandler:
self.function_registry.register_function("play_music")
self.function_registry.register_function("plugin_loader")
self.function_registry.register_function("get_time")
self.function_registry.register_function("raise_and_lower_the_volume")
def register_config_functions(self):
"""注册配置中的函数,可以不同客户端使用不同的配置"""
@@ -69,7 +70,7 @@ class FunctionHandler:
arguments = function_call_data["arguments"]
arguments = json.loads(arguments) if arguments else {}
logger.bind(tag=TAG).info(f"调用函数: {function_name}, 参数: {arguments}")
if funcItem.type == ToolType.SYSTEM_CTL:
if funcItem.type == ToolType.SYSTEM_CTL or funcItem.type == ToolType.IOT_CTL:
return func(conn, **arguments)
elif funcItem.type == ToolType.WAIT:
return func(**arguments)
@@ -1,5 +1,6 @@
from config.logger import setup_logging
import json
import uuid
from core.handle.sendAudioHandle import send_stt_message
from core.utils.util import remove_punctuation_and_length
@@ -8,30 +9,16 @@ logger = setup_logging()
async def handle_user_intent(conn, text):
"""
Handle user intent before starting chat
Args:
conn: Connection object
text: User's text input
Returns:
bool: True if intent was handled, False if should proceed to chat
"""
# 检查是否有明确的退出命令
if await check_direct_exit(conn, text):
return True
if conn.use_function_call_mode:
# 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
intent = await analyze_intent_with_llm(conn, text)
if not intent:
return False
# 处理各种意图
return await process_intent_result(conn, intent, text)
@@ -58,7 +45,6 @@ async def analyze_intent_with_llm(conn, text):
dialogue = conn.dialogue
try:
intent_result = await conn.intent.detect_intent(conn, dialogue.dialogue, text)
# 尝试解析JSON结果
try:
intent_data = json.loads(intent_result)
@@ -79,9 +65,6 @@ async def process_intent_result(conn, intent, original_text):
# 处理退出意图
if "结束聊天" in intent:
logger.bind(tag=TAG).info(f"识别到退出意图: {intent}")
# 如果正在播放音乐,可以关了 TODO
# 如果是明确的离别意图,发送告别语并关闭连接
await send_stt_message(conn, original_text)
conn.executor.submit(conn.chat_and_close, original_text)
@@ -90,10 +73,37 @@ async def process_intent_result(conn, intent, original_text):
# 处理播放音乐意图
if "播放音乐" in intent:
logger.bind(tag=TAG).info(f"识别到音乐播放意图: {intent}")
await conn.music_handler.handle_music_command(conn, intent)
# 调用play_music函数来播放音乐
song_name = extract_text_in_brackets(intent)
function_id = str(uuid.uuid4().hex)
function_name = "play_music"
function_arguments = '{ "song_name": "' + song_name + '" }'
function_call_data = {
"name": function_name,
"id": function_id,
"arguments": function_arguments
}
conn.func_handler.handle_llm_function_call(conn, function_call_data)
return True
# 其他意图处理可以在这里扩展
# 默认返回False,表示继续常规聊天流程
return False
def extract_text_in_brackets(s):
"""
从字符串中提取中括号内的文字
:param s: 输入字符串
:return: 中括号内的文字,如果不存在则返回空字符串
"""
left_bracket_index = s.find('[')
right_bracket_index = s.find(']')
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 ""
+273 -97
View File
@@ -1,24 +1,151 @@
import json
import asyncio
from config.logger import setup_logging
from plugins_func.register import device_type_registry, register_function, ActionResponse, Action, ToolType
TAG = __name__
logger = setup_logging()
def wrap_async_function(async_func):
"""包装异步函数为同步函数"""
def wrapper(*args, **kwargs):
try:
# 获取连接对象(第一个参数)
conn = args[0]
if not hasattr(conn, 'loop'):
logger.bind(tag=TAG).error("Connection对象没有loop属性")
return ActionResponse(Action.ERROR, "Connection对象没有loop属性",
"执行操作时出错: Connection对象没有loop属性")
# 使用conn对象中的事件循环
loop = conn.loop
# 在conn的事件循环中运行异步函数
future = asyncio.run_coroutine_threadsafe(async_func(*args, **kwargs), loop)
# 等待结果返回
return future.result()
except Exception as e:
logger.bind(tag=TAG).error(f"运行异步函数时出错: {e}")
return ActionResponse(Action.ERROR, str(e), f"执行操作时出错: {e}")
return wrapper
def create_iot_function(device_name, method_name, method_info):
"""
根据IOT设备描述生成通用的控制函数
"""
async def iot_control_function(conn, response_success=None, response_failure=None, **params):
try:
# 打印响应参数
logger.bind(tag=TAG).info(
f"控制函数接收到的响应参数: success='{response_success}', failure='{response_failure}'")
# 发送控制命令
await send_iot_conn(conn, device_name, method_name, params)
# 等待一小段时间让状态更新
await asyncio.sleep(0.1)
# 生成结果信息
result = f"{device_name}{method_name}操作执行成功"
# 根据方法名尝试自动更新状态
await update_state_by_method(conn, device_name, method_name, params)
# 处理响应中可能的占位符
response = response_success
# 替换{value}占位符
for param_name, param_value in params.items():
# 先尝试直接替换参数值
if "{" + param_name + "}" in response:
response = response.replace("{" + param_name + "}", str(param_value))
# 如果有{value}占位符,用相关参数替换
if "{value}" in response:
response = response.replace("{value}", str(param_value))
break
return ActionResponse(Action.RESPONSE, result, response)
except Exception as e:
logger.bind(tag=TAG).error(f"执行{device_name}{method_name}操作失败: {e}")
# 操作失败时使用大模型提供的失败响应
response = response_failure
return ActionResponse(Action.ERROR, str(e), response)
return wrap_async_function(iot_control_function)
def create_iot_query_function(device_name, prop_name, prop_info):
"""
根据IOT设备属性创建查询函数
"""
async def iot_query_function(conn, response_success=None, response_failure=None):
try:
# 打印响应参数
logger.bind(tag=TAG).info(
f"查询函数接收到的响应参数: success='{response_success}', failure='{response_failure}'")
value = await get_iot_status(conn, device_name, prop_name)
# 查询成功,生成结果
if value is not None:
# 使用大模型提供的成功响应,并替换其中的占位符
response = response_success.replace("{value}", str(value))
return ActionResponse(Action.RESPONSE, str(value), response)
else:
# 查询失败,使用大模型提供的失败响应
response = response_failure
return ActionResponse(Action.ERROR, f"属性{prop_name}不存在", response)
except Exception as e:
logger.bind(tag=TAG).error(f"查询{device_name}{prop_name}时出错: {e}")
# 查询出错时使用大模型提供的失败响应
response = response_failure
return ActionResponse(Action.ERROR, str(e), response)
return wrap_async_function(iot_query_function)
async def update_state_by_method(conn, device_name, method_name, params):
"""根据方法和参数自动更新设备状态"""
try:
# 规则1: 方法名为TurnOn,设置power为True
if method_name == "TurnOn":
await set_iot_status(conn, device_name, "power", True)
# 规则2: 方法名为TurnOff,设置power为False
elif method_name == "TurnOff":
await set_iot_status(conn, device_name, "power", False)
# 规则3: Set开头的方法,尝试更新对应参数
elif method_name.startswith("Set"):
# 从参数中找到可能的状态值
for param_name, param_value in params.items():
# 尝试更新对应名称的属性
await set_iot_status(conn, device_name, param_name, param_value)
# 其他方法,尝试直接从参数更新状态
else:
for param_name, param_value in params.items():
# 检查设备是否有此属性
status = await get_iot_status(conn, device_name, param_name)
if status is not None: # 属性存在
await set_iot_status(conn, device_name, param_name, param_value)
except Exception as e:
logger.bind(tag=TAG).warning(f"自动更新状态失败: {e}")
class IotDescriptor:
"""
A class to represent an IoT descriptor.
Attributes:
----------
name : str
The name of the IoT descriptor.
description : str
A brief description of the IoT descriptor.
properties : dict
A dictionary containing properties of the IoT descriptor.
methods : dict
A dictionary containing methods of the IoT descriptor.
-------
"""
def __init__(self, name, description, properties, methods):
@@ -29,17 +156,7 @@ class IotDescriptor:
# 根据描述创建属性
for key, value in properties.items():
# "volume":{"description":"当前音量 值","type":"number"}
"""
等价于
{
'name': 名字,
'description': 描述,
'value': 0
}
"""
# setattr(self, key, {}) # 创建一个空字典, 名字是属性名
property_item = globals()[key] = {} # 创建一个空字典, 名字是属性名
property_item = globals()[key] = {}
property_item['name'] = key
property_item["description"] = value["description"]
if value["type"] == "number":
@@ -52,23 +169,10 @@ class IotDescriptor:
# 根据描述创建方法
for key, value in methods.items():
# "SetVolume": {"description":"设置音量","parameters":{"volume":{"description":"0到100之间的整数","type":"number"}}}
"""
等价于
SetVolume = {
`description`: 描述,
`volume`: {
`description`: 描述,
`value`: 0
}
}
"""
# setattr(self, key, {}) # 创建一个空字典, 名字是方法名
method = globals()[key] = {} # 创建一个空字典, 名字是方法名
method = globals()[key] = {}
method["description"] = value["description"]
method['name'] = key
for k, v in value["parameters"].items():
# 不同的参数解析
method[k] = {}
method[k]["description"] = v["description"]
if v["type"] == "number":
@@ -77,58 +181,136 @@ class IotDescriptor:
method[k]["value"] = False
else:
method[k]["value"] = ""
self.methods.append(method)
async def handleIotDescriptors(conn, descriptors):
"""
处理物联网描述
示例: [{
"name":"Speaker",
"description":"当前 AI 机器人的扬声器",
"properties":{
"volume":{"description":"当前音量 值","type":"number"} 可以有boolean, number, string三种类型
},
"methods":{
"SetVolume":{
"description":"设置音量","parameters":{"volume":{"description":"0到100之间的整数","type":"number"}}
def register_device_type(descriptor):
"""注册设备类型及其功能"""
device_name = descriptor["name"]
type_id = device_type_registry.generate_device_type_id(descriptor)
# 如果该类型已注册,直接返回类型ID
if type_id in device_type_registry.type_functions:
return type_id
functions = {}
# 为每个属性创建查询函数
for prop_name, prop_info in descriptor["properties"].items():
func_name = f"get_{device_name.lower()}_{prop_name.lower()}"
func_desc = {
"type": "function",
"function": {
"name": func_name,
"description": f"查询{descriptor['description']}{prop_info['description']}",
"parameters": {
"type": "object",
"properties": {
"response_success": {
"type": "string",
"description": f"查询成功时的友好回复,必须使用{{value}}作为占位符表示查询到的值"
},
"response_failure": {
"type": "string",
"description": f"查询失败时的友好回复,例如:'无法获取{device_name}{prop_info['description']}'"
}
},
"required": ["response_success", "response_failure"]
}
}
}
}]
descriptors: 描述列表
"""
query_func = create_iot_query_function(device_name, prop_name, prop_info)
decorated_func = register_function(func_name, func_desc, ToolType.IOT_CTL)(query_func)
functions[func_name] = decorated_func
# 为每个方法创建控制函数
for method_name, method_info in descriptor["methods"].items():
func_name = f"{device_name.lower()}_{method_name.lower()}"
# 创建参数字典,添加原有参数
parameters = {
param_name: {
"type": param_info["type"],
"description": param_info["description"]
}
for param_name, param_info in method_info["parameters"].items()
}
# 添加响应参数
parameters.update({
"response_success": {
"type": "string",
"description": "操作成功时的友好回复,关于该设备的操作结果,设备名称尽量使用description中的名称"
},
"response_failure": {
"type": "string",
"description": "操作失败时的友好回复,关于该设备的操作结果,设备名称尽量使用description中的名称"
}
})
# 构建必须参数列表(原有参数 + 响应参数)
required_params = list(method_info["parameters"].keys())
required_params.extend(["response_success", "response_failure"])
func_desc = {
"type": "function",
"function": {
"name": func_name,
"description": f"{descriptor['description']} - {method_info['description']}",
"parameters": {
"type": "object",
"properties": parameters,
"required": required_params
}
}
}
control_func = create_iot_function(device_name, method_name, method_info)
decorated_func = register_function(func_name, func_desc, ToolType.IOT_CTL)(control_func)
functions[func_name] = decorated_func
device_type_registry.register_device_type(type_id, functions)
return type_id
# 用于接受前端设备推送的搜索iot描述
async def handleIotDescriptors(conn, descriptors):
"""处理物联网描述"""
functions_changed = False
for descriptor in descriptors:
# 创建IOT设备描述符
iot_descriptor = IotDescriptor(descriptor["name"], descriptor["description"], descriptor["properties"],
descriptor["methods"])
conn.iot_descriptors[descriptor["name"]] = iot_descriptor
# 暂时从配置文件中设置音量,后期通过意图识别控制音量
default_iot_volume = 100
if "iot" in conn.config:
default_iot_volume = conn.config["iot"]["Speaker"]["volume"]
logger.bind(tag=TAG).info(f"服务端设置音量为{default_iot_volume}")
await send_iot_conn(conn, "Speaker", "SetVolume", {"volume": default_iot_volume})
if conn.use_function_call_mode:
# 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)
# 在连接级注册设备函数
if hasattr(conn, 'func_handler'):
for func_name in device_functions:
conn.func_handler.function_registry.register_function(func_name)
logger.bind(tag=TAG).info(f"注册IOT函数到function handler: {func_name}")
functions_changed = True
# 如果注册了新函数,更新function描述列表
if functions_changed and hasattr(conn, 'func_handler'):
conn.func_handler.upload_functions_desc()
func_names = conn.func_handler.current_support_functions()
logger.bind(tag=TAG).info(f"设备类型: {type_id}")
logger.bind(tag=TAG).info(f"更新function描述列表完成,当前支持的函数: {func_names}")
async def handleIotStatus(conn, states):
"""
处理物联网状态
示例: [{
"name":"Speaker",
"state":{
"volume":100
}
}]
states: 状态列表
"""
"""处理物联网状态"""
for state in states:
for key, value in conn.iot_descriptors.items():
if key == state["name"]:
for property_item in value.properties:
# properties为字典列表, 记录各种属性
for k, v in state["state"].items():
# state为字典, 记录各种属性的值, 是需要记录的信息
if property_item["name"] == k:
# 检查一下属性是不是相同的
if type(v) != type(property_item["value"]):
logger.bind(tag=TAG).error(f"属性{property_item['name']}的值类型不匹配")
break
@@ -138,41 +320,35 @@ async def handleIotStatus(conn, states):
break
break
async def get_iot_status(conn, name, property_name):
"""
获取物联网状态
name: 设备名称 "Speaker"
property_name: 属性名称 "volume"
返回值: 属性值, 实际的属性有int, bool和str三种类型
"""
"""获取物联网状态"""
for key, value in conn.iot_descriptors.items():
if key == name:
for property_item in value.properties:
if property_item["name"] == property_name:
return property_item["value"]
logger.bind(tag=TAG).warning(f"未找到设备 {name} 的属性 {property_name}")
return None
async def send_iot_conn(conn, name, method_name, parameters):
"""
发送物联网指令
name: 设备名称 "Speaker"
method: 方法 "SetVolume"
parameters: 参数, 是一个字典 {"volume": 100}
发送示例:
{
"type": "iot",
"commands": [
{
"name" : "Speaker",
"method": "SetVolume",
"parameters": {
"volume": 100
}
}
]
}
"""
async def set_iot_status(conn, name, property_name, value):
"""设置物联网状态"""
for key, iot_descriptor in conn.iot_descriptors.items():
if key == name:
for property_item in iot_descriptor.properties:
if property_item["name"] == property_name:
if type(value) != type(property_item["value"]):
logger.bind(tag=TAG).error(f"属性{property_item['name']}的值类型不匹配")
return
property_item["value"] = value
logger.bind(tag=TAG).info(f"物联网状态更新: {name} , {property_name} = {value}")
return
logger.bind(tag=TAG).warning(f"未找到设备 {name} 的属性 {property_name}")
async def send_iot_conn(conn, name, method_name, parameters):
"""发送物联网指令"""
for key, value in conn.iot_descriptors.items():
if key == name:
# 找到了设备
@@ -1,139 +0,0 @@
from config.logger import setup_logging
import os
import random
import difflib
import re
import traceback
from pathlib import Path
import time
from core.handle.sendAudioHandle import send_stt_message
from core.utils import p3
TAG = __name__
logger = setup_logging()
def _extract_song_name(text):
"""从用户输入中提取歌名"""
for keyword in ["播放音乐"]:
if keyword in text:
parts = text.split(keyword)
if len(parts) > 1:
return parts[1].strip()
return None
def _find_best_match(potential_song, music_files):
"""查找最匹配的歌曲"""
best_match = None
highest_ratio = 0
for music_file in music_files:
song_name = os.path.splitext(music_file)[0]
ratio = difflib.SequenceMatcher(None, potential_song, song_name).ratio()
if ratio > highest_ratio and ratio > 0.4:
highest_ratio = ratio
best_match = music_file
return best_match
class MusicManager:
def __init__(self, music_dir, music_ext):
self.music_dir = Path(music_dir)
self.music_ext = music_ext
def get_music_files(self):
music_files = []
for file in self.music_dir.rglob("*"):
# 判断是否是文件
if file.is_file():
# 获取文件扩展名
ext = file.suffix.lower()
# 判断扩展名是否在列表中
if ext in self.music_ext:
# music_files.append(str(file.resolve())) # 添加绝对路径
# 添加相对路径
music_files.append(str(file.relative_to(self.music_dir)))
return music_files
class MusicHandler:
def __init__(self, config):
self.config = config
if "music" in self.config:
self.music_config = self.config["music"]
self.music_dir = os.path.abspath(
self.music_config.get("music_dir", "./music") # 默认路径修改
)
self.music_ext = self.music_config.get("music_ext", (".mp3", ".wav", ".p3"))
self.refresh_time = self.music_config.get("refresh_time", 60)
else:
self.music_dir = os.path.abspath("./music")
self.music_ext = (".mp3", ".wav", ".p3")
self.refresh_time = 60
# 获取音乐文件列表
self.music_files = MusicManager(self.music_dir, self.music_ext).get_music_files()
self.scan_time = time.time()
logger.bind(tag=TAG).debug(f"找到的音乐文件: {self.music_files}")
async def handle_music_command(self, conn, text):
"""处理音乐播放指令"""
clean_text = re.sub(r'[^\w\s]', '', text).strip()
logger.bind(tag=TAG).debug(f"检查是否是音乐命令: {clean_text}")
# 尝试匹配具体歌名
if os.path.exists(self.music_dir):
if time.time() - self.scan_time > self.refresh_time:
# 刷新音乐文件列表
self.music_files = MusicManager(self.music_dir, self.music_ext).get_music_files()
self.scan_time = time.time()
logger.bind(tag=TAG).debug(f"刷新的音乐文件: {self.music_files}")
potential_song = _extract_song_name(clean_text)
if potential_song:
best_match = _find_best_match(potential_song, self.music_files)
if best_match:
logger.bind(tag=TAG).info(f"找到最匹配的歌曲: {best_match}")
await self.play_local_music(conn, specific_file=best_match)
return True
# 检查是否是通用播放音乐命令
await self.play_local_music(conn)
return True
async def play_local_music(self, conn, specific_file=None):
"""播放本地音乐文件"""
try:
if not os.path.exists(self.music_dir):
logger.bind(tag=TAG).error(f"音乐目录不存在: {self.music_dir}")
return
# 确保路径正确性
if specific_file:
selected_music = specific_file
music_path = os.path.join(self.music_dir, specific_file)
else:
if not self.music_files:
logger.bind(tag=TAG).error("未找到MP3音乐文件")
return
selected_music = random.choice(self.music_files)
music_path = os.path.join(self.music_dir, selected_music)
if not os.path.exists(music_path):
logger.bind(tag=TAG).error(f"选定的音乐文件不存在: {music_path}")
return
text = f"正在播放{selected_music}"
await send_stt_message(conn, text)
conn.tts_first_text_index = 0
conn.tts_last_text_index = 0
conn.llm_finish_task = True
if music_path.endswith(".p3"):
opus_packets, duration = p3.decode_opus_from_file(music_path)
else:
opus_packets, duration = conn.tts.audio_to_opus_data(music_path)
conn.audio_play_queue.put((opus_packets, selected_music, 0))
except Exception as e:
logger.bind(tag=TAG).error(f"播放音乐失败: {str(e)}")
logger.bind(tag=TAG).error(f"详细错误: {traceback.format_exc()}")
@@ -2,7 +2,7 @@ from config.logger import setup_logging
import json
from core.handle.abortHandle import handleAbortMessage
from core.handle.helloHandle import handleHelloMessage
from core.handle.receiveAudioHandle import startToChat
from core.handle.receiveAudioHandle import startToChat, handleAudioMessage
from core.handle.iotHandle import handleIotDescriptors, handleIotStatus
TAG = __name__
@@ -31,6 +31,8 @@ async def handleTextMessage(conn, message):
elif msg_json["state"] == "stop":
conn.client_have_voice = True
conn.client_voice_stop = True
if len(conn.asr_audio) > 0:
await handleAudioMessage(conn, b'')
elif msg_json["state"] == "detect":
conn.asr_server_receive = False
conn.client_have_voice = False
@@ -41,6 +43,6 @@ async def handleTextMessage(conn, message):
if "descriptors" in msg_json:
await handleIotDescriptors(conn, msg_json["descriptors"])
if "states" in msg_json:
await handleIotStatus(conn, msg_json["states"])
await handleIotStatus(conn, msg_json["states"])
except json.JSONDecodeError:
await conn.websocket.send(message)
@@ -1,12 +1,13 @@
from typing import List, Dict
from ..base import IntentProviderBase
from plugins_func.functions.play_music import initialize_music_handler
from config.logger import setup_logging
import re
import re
TAG = __name__
logger = setup_logging()
class IntentProvider(IntentProviderBase):
def __init__(self, config):
super().__init__(config)
@@ -73,8 +74,8 @@ class IntentProvider(IntentProviderBase):
"你现在可以使用的音乐的名称如下(使用<start>和<end>标志):\n"
)
return prompt
async def detect_intent(self, conn, dialogue_history: List[Dict], text:str) -> str:
async def detect_intent(self, conn, dialogue_history: List[Dict], text: str) -> str:
if not self.llm:
raise ValueError("LLM provider not set")
@@ -89,7 +90,9 @@ class IntentProvider(IntentProviderBase):
msgStr += f"User: {text}\n"
user_prompt = f"当前的对话如下:\n{msgStr}"
prompt_music = f"{self.promot}\n<start>{conn.music_handler.music_files}\n<end>"
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>"
logger.bind(tag=TAG).debug(f"User prompt: {prompt_music}")
# 使用LLM进行意图识别
intent = self.llm.response_no_stream(
@@ -100,10 +103,9 @@ class IntentProvider(IntentProviderBase):
# 使用正则表达式提取 {} 中的内容
match = re.search(r'\{.*?\}', intent)
if match:
result = match.group(0) # 获取匹配到的内容(包含 {}
print(result) # 输出:{intent: '播放音乐 [中秋月]'}
result = match.group(0)
intent = result
else:
intent = "{intent: '继续聊天'}"
logger.bind(tag=TAG).info(f"Detected intent: {intent}")
return intent.strip()
return intent.strip()
@@ -12,10 +12,10 @@ class LLMProvider(LLMProviderBase):
self.model_name = config.get("model_name")
self.base_url = config.get("base_url", "http://localhost:11434")
# Initialize OpenAI client with Ollama base URL
#如果没有v1,增加v1
# 如果没有v1,增加v1
if not self.base_url.endswith("/v1"):
self.base_url = f"{self.base_url}/v1"
self.client = OpenAI(
base_url=self.base_url,
api_key="ollama" # Ollama doesn't need an API key but OpenAI client requires one
@@ -28,13 +28,20 @@ class LLMProvider(LLMProviderBase):
messages=dialogue,
stream=True
)
is_active=True
for chunk in responses:
try:
delta = chunk.choices[0].delta if getattr(chunk, 'choices', None) else None
content = delta.content if hasattr(delta, 'content') else ''
if content:
yield content
if '<think>' in content:
is_active = False
content = content.split('<think>')[0]
if '</think>' in content:
is_active = True
content = content.split('</think>')[-1]
if is_active:
yield content
except Exception as e:
logger.bind(tag=TAG).error(f"Error processing chunk: {e}")
@@ -50,10 +57,10 @@ class LLMProvider(LLMProviderBase):
stream=True,
tools=functions,
)
for chunk in stream:
yield chunk.choices[0].delta.content, chunk.choices[0].delta.tool_calls
except Exception as e:
logger.bind(tag=TAG).error(f"Error in Ollama function call: {e}")
yield {"type": "content", "content": f"【Ollama服务响应异常: {str(e)}"}
yield {"type": "content", "content": f"【Ollama服务响应异常: {str(e)}"}
@@ -13,7 +13,7 @@ logger = setup_logging()
class TTSProviderBase(ABC):
def __init__(self, config, delete_audio_file):
self.delete_audio_file = delete_audio_file
self.output_file = config.get("output_file")
self.output_file = config.get("output_dir")
@abstractmethod
def generate_filename(self):
@@ -15,7 +15,7 @@ class TTSProvider(TTSProviderBase):
self.headers = config.get("headers", {})
self.params = config.get("params")
self.format = config.get("format", "wav")
self.output_file = config.get("output_file", "tmp/")
self.output_file = config.get("output_dir", "tmp/")
def generate_filename(self):
return os.path.join(self.output_file, f"tts-{datetime.now().date()}@{uuid.uuid4().hex}.{self.format}")
@@ -12,17 +12,18 @@ class TTSProvider(TTSProviderBase):
def __init__(self, config, delete_audio_file):
super().__init__(config, delete_audio_file)
self.url = config.get("url")
self.text_lang = config.get("text_lang", "audo")
self.ref_audio_path = config.get("ref_audio_path")
self.prompt_lang = config.get("prompt_lang")
self.refer_wav_path = config.get("refer_wav_path")
self.prompt_text = config.get("prompt_text")
self.top_k = config.get("top_k", 5)
self.top_p = config.get("top_p", 1)
self.temperature = config.get("temperature", 1)
self.sample_steps = config.get("sample_steps", 16)
self.media_type = config.get("media_type", "wav")
self.streaming_mode = config.get("streaming_mode", False)
self.threshold = config.get("threshold", 30)
self.prompt_language = config.get("prompt_language")
self.text_language = config.get("text_language", "audo")
self.top_k = config.get("top_k", 15)
self.top_p = config.get("top_p", 1.0)
self.temperature = config.get("temperature", 1.0)
self.cut_punc = config.get("cut_punc","")
self.speed = config.get("speed", 1.0)
self.inp_refs = config.get("inp_refs",[])
self.sample_steps = config.get("sample_steps",32)
self.if_sr = config.get("if_sr",False)
def generate_filename(self, extension=".wav"):
@@ -30,18 +31,19 @@ class TTSProvider(TTSProviderBase):
async def text_to_speak(self, text, output_file):
request_params = {
"text": text,
"text_lang": self.text_lang,
"ref_audio_path": self.ref_audio_path,
"prompt_lang": self.prompt_lang,
"refer_wav_path": self.refer_wav_path,
"prompt_text": self.prompt_text,
"prompt_language": self.prompt_language,
"text": text,
"text_language": self.text_language,
"top_k": self.top_k,
"top_p": self.top_p,
"temperature": self.temperature,
"cut_punc": self.cut_punc,
"speed": self.speed,
"inp_refs": self.inp_refs,
"sample_steps": self.sample_steps,
"media_type": self.media_type,
"streaming_mode": self.streaming_mode,
"threshold": self.threshold,
"if_sr": self.if_sr,
}
resp = requests.get(self.url, params=request_params)
@@ -14,7 +14,7 @@ class TTSProvider(TTSProviderBase):
self.voice = config.get("voice", "alloy")
self.response_format = "wav"
self.speed = config.get("speed", 1.0)
self.output_file = config.get("output_file", "tmp/")
self.output_file = config.get("output_dir", "tmp/")
check_model_key("TTS", self.api_key)
def generate_filename(self, extension=".wav"):
@@ -17,7 +17,7 @@ class TTSProvider(TTSProviderBase):
self.volume_change_dB = config.get("volume_change_dB", 0)
self.speed_factor = config.get("speed_factor", 1)
self.stream = config.get("stream", False)
self.output_file = config.get("output_file")
self.output_file = config.get("output_dir")
self.pitch_factor = config.get("pitch_factor", 0)
self.format = config.get("format", "mp3")
self.emotion = config.get("emotion", 1)
+59 -1
View File
@@ -1,11 +1,11 @@
import os
import json
from datetime import datetime
import yaml
import socket
import subprocess
import logging
import re
import requests
def get_project_dir():
@@ -24,6 +24,64 @@ def get_local_ip():
except Exception as e:
return "127.0.0.1"
def is_private_ip(ip_addr):
"""
Check if an IP address is a private IP address (compatible with IPv4 and IPv6).
@param {string} ip_addr - The IP address to check.
@return {bool} True if the IP address is private, False otherwise.
"""
try:
# Validate IPv4 or IPv6 address format
if not re.match(r"^(\d{1,3}\.){3}\d{1,3}$|^([0-9a-fA-F]{1,4}:){7}[0-9a-fA-F]{1,4}$", ip_addr):
return False # Invalid IP address format
# IPv4 private address ranges
if '.' in ip_addr: # IPv4 address
ip_parts = list(map(int, ip_addr.split('.')))
if ip_parts[0] == 10:
return True # 10.0.0.0/8 range
elif ip_parts[0] == 172 and 16 <= ip_parts[1] <= 31:
return True # 172.16.0.0/12 range
elif ip_parts[0] == 192 and ip_parts[1] == 168:
return True # 192.168.0.0/16 range
elif ip_addr == '127.0.0.1':
return True # Loopback address
elif ip_parts[0] == 169 and ip_parts[1] == 254:
return True # Link-local address 169.254.0.0/16
else:
return False # Not a private IPv4 address
else: # IPv6 address
ip_addr = ip_addr.lower()
if ip_addr.startswith('fc00:') or ip_addr.startswith('fd00:'):
return True # Unique Local Addresses (FC00::/7)
elif ip_addr == '::1':
return True # Loopback address
elif ip_addr.startswith('fe80:'):
return True # Link-local unicast addresses (FE80::/10)
else:
return False # Not a private IPv6 address
except (ValueError, IndexError):
return False # IP address format error or insufficient segments
def get_ip_info(ip_addr):
try:
base_url = "https://freeipapi.com/api/json"
url = base_url if is_private_ip(ip_addr) else f"{base_url}/{ip_addr}"
resp = requests.get(url).json()
ip_info = {
"city": resp.get("cityName"),
"region": resp.get("regionName"),
"country": resp.get("countryName")
}
return ip_info
except Exception as e:
logging.error(f"Error getting client ip info: {e}")
return {}
def read_config(config_path):
with open(config_path, "r", encoding="utf-8") as file:
+2 -4
View File
@@ -2,7 +2,6 @@ import asyncio
import websockets
from config.logger import setup_logging
from core.connection import ConnectionHandler
from core.handle.musicHandler import MusicHandler
from core.utils.util import get_local_ip
from core.utils import asr, vad, llm, tts, memory, intent
@@ -13,7 +12,7 @@ class WebSocketServer:
def __init__(self, config: dict):
self.config = config
self.logger = setup_logging()
self._vad, self._asr, self._llm, self._tts, self._music, self._memory, self.intent = self._create_processing_instances()
self._vad, self._asr, self._llm, self._tts, self._memory, self.intent = self._create_processing_instances()
self.active_connections = set() # 添加全局连接记录
def _create_processing_instances(self):
@@ -50,7 +49,6 @@ class WebSocketServer:
self.config["TTS"][self.config["selected_module"]["TTS"]],
self.config["delete_audio"]
),
MusicHandler(self.config),
memory.create_instance(memory_cls_name, memory_cfg),
intent.create_instance(
self.config["selected_module"]["Intent"]
@@ -80,7 +78,7 @@ class WebSocketServer:
async def _handle_connection(self, websocket):
"""处理新连接,每次创建独立的ConnectionHandler"""
# 创建ConnectionHandler时传入当前server实例
handler = ConnectionHandler(self.config, self._vad, self._asr, self._llm, self._tts, self._music, self._memory, self.intent)
handler = ConnectionHandler(self.config, self._vad, self._asr, self._llm, self._tts, self._memory, self.intent)
self.active_connections.add(handler)
try:
await handler.handle_connection(websocket)
+1
View File
@@ -21,6 +21,7 @@ services:
- ./data:/opt/xiaozhi-esp32-server/data
# 模型文件挂接,很重要
- ./models/SenseVoiceSmall/model.pt:/opt/xiaozhi-esp32-server/models/SenseVoiceSmall/model.pt
# #智控台还没开发好,还不能完全使用,会报很多错误,如果是非技术人员,请不要启用智控台服务
# xiaozhi-esp32-server-web:
# image: ghcr.nju.edu.cn/xinnan-tech/xiaozhi-esp32-server:web_latest
@@ -1,42 +1,104 @@
import requests
from bs4 import BeautifulSoup
from plugins_func.register import register_function,ToolType, ActionResponse, Action
from config.logger import setup_logging
from plugins_func.register import register_function, ToolType, ActionResponse, Action
TAG = __name__
logger = setup_logging()
get_weather_function_desc = {
GET_WEATHER_FUNCTION_DESC = {
"type": "function",
"function": {
"name": "get_weather",
"description": "获取某个地点的天气,用户应先提供一个位置,比如用户说杭州天气,参数为:zhejiang/hangzhou,比如用户说北京天气怎么样,参数为:beijing/beijing。如果用户只问天气怎么样,参数是:guangdong/guangzhou",
"description": (
"获取某个地点的天气,用户应提供一个位置,比如用户说杭州天气,参数为:杭州。"
"如果用户说的是省份,默认用省会城市。如果用户说的不是省份或城市而是一个地名,"
"默认用该地所在省份的省会城市。"
),
"parameters": {
"type": "object",
"properties": {
"city": {
"location": {
"type": "string",
"description": "城市,zhejiang/hangzhou"
"description": "地点名,例如杭州。可选参数,如果不提供则不传"
},
"lang": {
"type": "string",
"description": "返回用户使用的语言code,例如zh_CN/zh_HK/en_US/ja_JP等,默认zh_CN"
}
},
"required": [
"city"
]
"required": ["lang"]
}
}
}
@register_function('get_weather', get_weather_function_desc, ToolType.WAIT)
def get_weather(city: str):
"""
"获取某个地点的天气,用户应先提供一个位置,\n比如用户说杭州天气,参数为:zhejiang/hangzhou\n\n比如用户说北京天气怎么样,参数为:beijing/beijing",
city : 城市,zhejiang/hangzhou
"""
url = "https://tianqi.moji.com/weather/china/"+city
headers = {
'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/92.0.4515.107 Safari/537.36'
}
response = requests.get(url, headers=headers)
if response.status_code!=200:
HEADERS = {
'User-Agent': (
'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 '
'(KHTML, like Gecko) Chrome/92.0.4515.107 Safari/537.36'
)
}
def fetch_city_info(location, api_key):
url = f"https://geoapi.qweather.com/v2/city/lookup?key={api_key}&location={location}&lang=zh"
response = requests.get(url, headers=HEADERS).json()
return response.get('location', [])[0] if response.get('location') else None
def fetch_weather_page(url):
response = requests.get(url, headers=HEADERS)
return BeautifulSoup(response.text, "html.parser") if response.ok else None
def parse_weather_info(soup):
city_name = soup.select_one("h1.c-submenu__location").get_text(strip=True)
current_abstract = soup.select_one(".c-city-weather-current .current-abstract")
current_abstract = current_abstract.get_text(strip=True) if current_abstract else "未知"
current_basic = {}
for item in soup.select(".c-city-weather-current .current-basic .current-basic___item"):
parts = item.get_text(strip=True, separator=" ").split(" ")
if len(parts) == 2:
key, value = parts[1], parts[0]
current_basic[key] = value
temps_list = []
for row in soup.select(".city-forecast-tabs__row")[:7]: # 取前7天的数据
date = row.select_one(".date-bg .date").get_text(strip=True)
temps = [span.get_text(strip=True) for span in row.select(".tmp-cont .temp")]
high_temp, low_temp = (temps[0], temps[-1]) if len(temps) >= 2 else (None, None)
temps_list.append((date, high_temp, low_temp))
return city_name, current_abstract, current_basic, temps_list
@register_function('get_weather', GET_WEATHER_FUNCTION_DESC, ToolType.SYSTEM_CTL)
def get_weather(conn, location: str = None, lang: str = "zh_CN"):
api_key = conn.config["plugins"]["get_weather"]["api_key"]
default_location = conn.config["plugins"]["get_weather"]["default_location"]
location = location or conn.client_ip_info.get("city") or default_location
logger.bind(tag=TAG).debug(f"获取天气: {location}")
city_info = fetch_city_info(location, api_key)
if not city_info:
return ActionResponse(Action.REQLLM, f"未找到相关的城市: {location},请确认地点是否正确", None)
soup = fetch_weather_page(city_info['fxLink'])
if not soup:
return ActionResponse(Action.REQLLM, None, "请求失败")
soup = BeautifulSoup(response.text, "html.parser")
weather = soup.find('meta', attrs={'name':'description'})["content"]
weather = weather.replace("墨迹天气", "")
return ActionResponse(Action.REQLLM, weather, None)
city_name, current_abstract, current_basic, temps_list = parse_weather_info(soup)
weather_report = f"根据下列数据,用{lang}回应用户的查询天气请求:\n{city_name}未来7天天气:\n"
for i, (date, high, low) in enumerate(temps_list):
if high and low:
weather_report += f"{date}: {low}{high}\n"
weather_report += (
f"当前天气: {current_abstract}\n"
f"当前天气参数: {current_basic}\n"
f"(确保只报告指定单日的气温范围,除非用户明确要求想要了解多日天气,如果未指定,默认报告今天的温度范围。"
"参数为0的值不需要报告给用户,每次都报告体感温度,根据语境选择合适的参数内容告知用户,并对参数给出相应评价)"
)
return ActionResponse(Action.REQLLM, weather_report, None)
@@ -1,10 +1,22 @@
from plugins_func.register import register_function,ToolType, ActionResponse, Action
from config.logger import setup_logging
import os
import re
import time
import random
import asyncio
import difflib
import traceback
from pathlib import Path
from core.utils import p3
from core.handle.sendAudioHandle import send_stt_message
from plugins_func.register import register_function,ToolType, ActionResponse, Action
TAG = __name__
logger = setup_logging()
MUSIC_CACHE = {}
play_music_function_desc = {
"type": "function",
"function": {
@@ -29,12 +41,156 @@ def play_music(conn, song_name: str):
try:
music_intent = f"播放音乐 {song_name}" if song_name != "random" else "随机播放音乐"
# 执行音乐播放命令
# 检查事件循环状态
if not conn.loop.is_running():
logger.bind(tag=TAG).error("事件循环未运行,无法提交任务")
return ActionResponse(action=Action.RESPONSE, result="系统繁忙", response="请稍后再试")
# 提交异步任务
future = asyncio.run_coroutine_threadsafe(
conn.music_handler.handle_music_command(conn, music_intent),
handle_music_command(conn, music_intent),
conn.loop
)
future.result()
return ActionResponse(action=Action.RESPONSE, result="退出意图已处理", response="还想听什么歌?")
# 非阻塞回调处理
def handle_done(f):
try:
f.result() # 可在此处理成功逻辑
logger.bind(tag=TAG).info("播放完成")
except Exception as e:
logger.bind(tag=TAG).error(f"播放失败: {e}")
future.add_done_callback(handle_done)
return ActionResponse(action=Action.RESPONSE, result="指令已接收", response="正在为您播放音乐")
except Exception as e:
logger.bind(tag=TAG).error(f"处理音乐意图错误: {e}")
logger.bind(tag=TAG).error(f"处理音乐意图错误: {e}")
return ActionResponse(action=Action.RESPONSE, result=str(e), response="播放音乐时出错了")
def _extract_song_name(text):
"""从用户输入中提取歌名"""
for keyword in ["播放音乐"]:
if keyword in text:
parts = text.split(keyword)
if len(parts) > 1:
return parts[1].strip()
return None
def _find_best_match(potential_song, music_files):
"""查找最匹配的歌曲"""
best_match = None
highest_ratio = 0
for music_file in music_files:
song_name = os.path.splitext(music_file)[0]
ratio = difflib.SequenceMatcher(None, potential_song, song_name).ratio()
if ratio > highest_ratio and ratio > 0.4:
highest_ratio = ratio
best_match = music_file
return best_match
def get_music_files(music_dir, music_ext):
music_dir = Path(music_dir)
music_files = []
music_file_names = []
for file in music_dir.rglob("*"):
# 判断是否是文件
if file.is_file():
# 获取文件扩展名
ext = file.suffix.lower()
# 判断扩展名是否在列表中
if ext in music_ext:
# 添加相对路径
music_files.append(str(file.relative_to(music_dir)))
music_file_names.append(os.path.splitext(str(file.relative_to(music_dir)))[0])
return music_files, music_file_names
def initialize_music_handler(conn):
global MUSIC_CACHE
if MUSIC_CACHE == {}:
if "music" in conn.config:
MUSIC_CACHE["music_config"] = conn.config["music"]
MUSIC_CACHE["music_dir"] = os.path.abspath(
MUSIC_CACHE["music_config"].get("music_dir", "./music") # 默认路径修改
)
MUSIC_CACHE["music_ext"] = MUSIC_CACHE["music_config"].get("music_ext", (".mp3", ".wav", ".p3"))
MUSIC_CACHE["refresh_time"] = MUSIC_CACHE["music_config"].get("refresh_time", 60)
else:
MUSIC_CACHE["music_dir"] = os.path.abspath("./music")
MUSIC_CACHE["music_ext"] = (".mp3", ".wav", ".p3")
MUSIC_CACHE["refresh_time"] = 60
# 获取音乐文件列表
MUSIC_CACHE["music_files"], MUSIC_CACHE["music_file_names"] = get_music_files(MUSIC_CACHE["music_dir"],
MUSIC_CACHE["music_ext"])
MUSIC_CACHE["scan_time"] = time.time()
return MUSIC_CACHE
async def handle_music_command(conn, text):
initialize_music_handler(conn)
global MUSIC_CACHE
"""处理音乐播放指令"""
clean_text = re.sub(r'[^\w\s]', '', text).strip()
logger.bind(tag=TAG).debug(f"检查是否是音乐命令: {clean_text}")
# 尝试匹配具体歌名
if os.path.exists(MUSIC_CACHE["music_dir"]):
if time.time() - MUSIC_CACHE["scan_time"] > MUSIC_CACHE["refresh_time"]:
# 刷新音乐文件列表
MUSIC_CACHE["music_files"], MUSIC_CACHE["music_file_names"] = get_music_files(MUSIC_CACHE["music_dir"],
MUSIC_CACHE["music_ext"])
MUSIC_CACHE["scan_time"] = time.time()
potential_song = _extract_song_name(clean_text)
if potential_song:
best_match = _find_best_match(potential_song, MUSIC_CACHE["music_files"])
if best_match:
logger.bind(tag=TAG).info(f"找到最匹配的歌曲: {best_match}")
await play_local_music(conn, specific_file=best_match)
return True
# 检查是否是通用播放音乐命令
await play_local_music(conn)
return True
async def play_local_music(conn, specific_file=None):
global MUSIC_CACHE
"""播放本地音乐文件"""
try:
if not os.path.exists(MUSIC_CACHE["music_dir"]):
logger.bind(tag=TAG).error(f"音乐目录不存在: " + MUSIC_CACHE["music_dir"])
return
# 确保路径正确性
if specific_file:
selected_music = specific_file
music_path = os.path.join(MUSIC_CACHE["music_dir"], specific_file)
else:
if not MUSIC_CACHE["music_files"]:
logger.bind(tag=TAG).error("未找到MP3音乐文件")
return
selected_music = random.choice(MUSIC_CACHE["music_files"])
music_path = os.path.join(MUSIC_CACHE["music_dir"], selected_music)
if not os.path.exists(music_path):
logger.bind(tag=TAG).error(f"选定的音乐文件不存在: {music_path}")
return
text = f"正在播放{selected_music}"
await send_stt_message(conn, text)
conn.tts_first_text_index = 0
conn.tts_last_text_index = 0
conn.llm_finish_task = True
if music_path.endswith(".p3"):
opus_packets, duration = p3.decode_opus_from_file(music_path)
else:
opus_packets, duration = conn.tts.audio_to_opus_data(music_path)
conn.audio_play_queue.put((opus_packets, selected_music, 0))
except Exception as e:
logger.bind(tag=TAG).error(f"播放音乐失败: {str(e)}")
logger.bind(tag=TAG).error(f"详细错误: {traceback.format_exc()}")
@@ -0,0 +1,64 @@
from config.logger import setup_logging
from plugins_func.register import register_function, ToolType, ActionResponse, Action
from core.handle.iotHandle import get_iot_status, send_iot_conn
import asyncio
TAG = __name__
logger = setup_logging()
raise_and_lower_the_volume_function_desc = {
"type": "function",
"function": {
"name": "raise_and_lower_the_volume",
"description": "用户觉得声音过高或过低,或者用户想提高或降低音量。比如用户说太大声了,参数为:lower,比如用户说提高音量,参数为:raise",
"parameters": {
"type": "object",
"properties": {
"action": {
"type": "string",
"description": "动作名称,要么是raise,要么是lower"
}
},
"required": ["action"]
}
}
}
@register_function('raise_and_lower_the_volume', raise_and_lower_the_volume_function_desc, ToolType.IOT_CTL)
def raise_and_lower_the_volume(conn, action: str):
"""
获取当前设备音量
"""
future = asyncio.run_coroutine_threadsafe(
_raise_and_lower_the_volume(conn, action),
conn.loop
)
try:
new_volume = future.result() # 同步等待异步操作完成
logger.bind(tag=TAG).info(f"音量操作完成: {new_volume}")
response = f"音量已调整到{new_volume}"
except Exception as e:
logger.bind(tag=TAG).error(f"音量操作失败: {e}")
response = f"音量调整失败: {e}"
return ActionResponse(action=Action.RESPONSE, result="指令已接收", response=response)
async def _raise_and_lower_the_volume(conn, action):
volume = await get_iot_status(conn, "Speaker", "volume")
if volume is None:
raise Exception("你的设备不支持音量控制")
if action == 'raise':
volume += 10
elif action == 'lower':
volume -= 10
# 限制音量范围在0到100之间
if volume < 0:
volume = 0
elif volume > 100:
volume = 100
await send_iot_conn(conn, "Speaker", "SetVolume", {"volume": volume})
return volume
@@ -11,6 +11,7 @@ class ToolType(Enum):
WAIT = (2, "调用工具,等待函数返回")
CHANGE_SYS_PROMPT = (3, "修改系统提示词,切换角色性格或职责")
SYSTEM_CTL = (4, "系统控制,影响正常的对话流程,如退出、播放音乐等,需要传递conn参数")
IOT_CTL = (5, "IOT设备控制,需要传递conn参数")
def __init__(self, code, message):
self.code = code
@@ -18,6 +19,7 @@ class ToolType(Enum):
class Action(Enum):
ERROR = (-1, "错误")
NOTFOUND = (0, "没有找到函数")
NONE = (1, "啥也不干")
RESPONSE = (2, "直接回复")
@@ -40,8 +42,31 @@ class FunctionItem:
self.func = func
self.type = type
class DeviceTypeRegistry:
"""设备类型注册表,用于管理IOT设备类型及其函数"""
def __init__(self):
self.type_functions = {} # type_signature -> {func_name: FunctionItem}
def generate_device_type_id(self, descriptor):
"""通过设备能力描述生成类型ID"""
properties = sorted(descriptor["properties"].keys())
methods = sorted(descriptor["methods"].keys())
# 使用属性和方法的组合作为设备类型的唯一标识
type_signature = f"{descriptor['name']}:{','.join(properties)}:{','.join(methods)}"
return type_signature
def get_device_functions(self, type_id):
"""获取设备类型对应的所有函数"""
return self.type_functions.get(type_id, {})
def register_device_type(self, type_id, functions):
"""注册设备类型及其函数"""
if type_id not in self.type_functions:
self.type_functions[type_id] = functions
# 初始化函数注册字典
all_function_registry = {}
device_type_registry = DeviceTypeRegistry()
def register_function(name, desc, type=None):
"""注册函数到函数注册字典的装饰器"""
+1 -1
View File
@@ -19,4 +19,4 @@ loguru==0.7.3
requests==2.32.3
cozepy==0.12.0
mem0ai==0.1.62
bs4==0.0.2
bs4==0.0.2