Merge pull request #1976 from xinnan-tech/docker-install

Docker install

README.md

@@ -235,7 +235,7 @@ Websocket接口地址: wss://2662r3426b.vicp.fun/xiaozhi/v1/
 ---
 ## 功能清单 ✨
 ### 已实现 ✅
-
+![请参考-全模块安装架构图](docs/images/deploy2.png)
 | 功能模块 | 描述 |
 |:---:|:---|
 | 核心架构 | 基于WebSocket和HTTP服务器，提供完整的控制台管理和认证系统 |
@@ -271,7 +271,6 @@ Websocket接口地址: wss://2662r3426b.vicp.fun/xiaozhi/v1/
 ---
 
 ## 本项目支持的平台/组件列表 📋
-
 ### LLM 语言模型
 
 | 使用方式 | 支持平台 | 免费平台 |

README_en.md

@@ -233,7 +233,7 @@ This project provides the following testing tools to help you verify the system
 ---
 ## Feature List ✨
 ### Implemented ✅
-
+![请参考-全模块安装架构图](docs/images/deploy2.png)
 | Feature Module | Description |
 |:---:|:---|
 | Core Architecture | Based on WebSocket and HTTP servers, provides complete console management and authentication system |
@@ -269,7 +269,6 @@ Xiaozhi is an ecosystem. When using this product, you can also check out other e
 ---
 
 ## Supported Platforms/Components List 📋
-
 ### LLM Language Models
 
 | Usage Method | Supported Platforms | Free Platforms |

docker-setup.sh

@@ -1,105 +1,414 @@
 #!/bin/sh
+# 脚本作者@VanillaNahida
 # 本文件是用于一键自动下载本项目所需文件，自动创建好目录
-# 所需条件（否则无法使用）：
+# 暂且只支持X86版本的Ubuntu系统，其他系统未测试
-# 1、请确保你的环境可以正常访问 GitHub 否则无法下载脚本
+
-#
+# 定义中断处理函数
-# 检测操作系统类型
+handle_interrupt() {
-case "$(uname -s)" in
+    echo ""
-    Linux*)     OS=Linux;;
+    echo "安装已被用户中断(Ctrl+C或Esc)"
-    Darwin*)    OS=Mac;;
+    echo "如需重新安装，请再次运行脚本"
-    CYGWIN*)    OS=Windows;;
+    exit 1
-    MINGW*)     OS=Windows;;
+}
-    MSYS*)      OS=Windows;;
+
-    *)          OS=UNKNOWN;;
+# 设置信号捕获，处理Ctrl+C
+trap handle_interrupt SIGINT
+
+# 处理Esc键
+# 保存终端设置
+old_stty_settings=$(stty -g)
+# 设置终端立即响应，不回显
+stty -icanon -echo min 1 time 0
+
+# 后台进程检测Esc键
+(while true; do
+    read -r key
+    if [[ $key == $'\e' ]]; then
+        # 检测到Esc键，触发中断处理
+        kill -SIGINT $$
+        break
+    fi
+done) &
+
+# 脚本结束时恢复终端设置
+trap 'stty "$old_stty_settings"' EXIT
+
+
+# 打印彩色字符画
+echo -e "\e[1;32m"  # 设置颜色为亮绿色
+cat << "EOF"
+脚本作者：@Bilibili 香草味的纳西妲喵
+ __      __            _  _  _            _   _         _      _      _        
+ \ \    / /           (_)| || |          | \ | |       | |    (_)    | |       
+  \ \  / /__ _  _ __   _ | || |  __ _    |  \| |  __ _ | |__   _   __| |  __ _ 
+   \ \/ // _` || '_ \ | || || | / _` |   | . ` | / _` || '_ \ | | / _` | / _` |
+    \  /| (_| || | | || || || || (_| |   | |\  || (_| || | | || || (_| || (_| |
+     \/  \__,_||_| |_||_||_||_| \__,_|   |_| \_| \__,_||_| |_||_| \__,_| \__,_|                                                                                                                                                                                                                               
+EOF
+echo -e "\e[0m"  # 重置颜色
+echo -e "\e[1;36m  小智服务端全量部署一键安装脚本 Ver 0.2 \e[0m\n"
+sleep 1
+
+
+
+# 检查并安装whiptail
+check_whiptail() {
+    if ! command -v whiptail &> /dev/null; then
+        echo "正在安装whiptail..."
+        apt update
+        apt install -y whiptail
+    fi
+}
+
+check_whiptail
+
+# 创建确认对话框
+whiptail --title "安装确认" --yesno "即将安装小智服务端，是否继续？" \
+  --yes-button "继续" --no-button "退出" 10 50
+
+# 根据用户选择执行操作
+case $? in
+  0)
+    ;;
+  1)
+    exit 1
+    ;;
 esac
 
-# 设置颜色（Windows CMD 不支持，但不影响使用）
+# 检查root权限
-if [ "$OS" = "Windows" ]; then
+if [ $EUID -ne 0 ]; then
-    GREEN=""
+    whiptail --title "权限错误" --msgbox "请使用root权限运行本脚本" 10 50
-    RED=""
-    NC=""
-else
-    GREEN='\033[0;32m'
-    RED='\033[0;31m'
-    NC='\033[0m'
-fi
-
-echo "${GREEN}开始安装小智服务端...${NC}"
-
-# 创建必要的目录
-echo "创建目录结构..."
-mkdir -p xiaozhi-server/data xiaozhi-server/models/SenseVoiceSmall
-cd xiaozhi-server || exit
-
-# 根据操作系统选择下载命令
-if [ "$OS" = "Windows" ]; then
-    DOWNLOAD_CMD="curl -L -o"
-    if ! command -v curl >/dev/null 2>&1; then
-        DOWNLOAD_CMD="powershell -Command Invoke-WebRequest -Uri"
-        DOWNLOAD_CMD_SUFFIX="-OutFile"
-    fi
-else
-    if command -v curl >/dev/null 2>&1; then
-        DOWNLOAD_CMD="curl -L -o"
-    elif command -v wget >/dev/null 2>&1; then
-        DOWNLOAD_CMD="wget -O"
-    else
-        echo "${RED}错误: 需要安装 curl 或 wget${NC}"
-        exit 1
-    fi
-fi
-
-# 下载语音识别模型
-echo "下载语音识别模型..."
-if [ "$DOWNLOAD_CMD" = "powershell -Command Invoke-WebRequest -Uri" ]; then
-    $DOWNLOAD_CMD "https://modelscope.cn/models/iic/SenseVoiceSmall/resolve/master/model.pt" $DOWNLOAD_CMD_SUFFIX "models/SenseVoiceSmall/model.pt"
-else
-    $DOWNLOAD_CMD "models/SenseVoiceSmall/model.pt" "https://modelscope.cn/models/iic/SenseVoiceSmall/resolve/master/model.pt"
-fi
-
-if [ $? -ne 0 ]; then
-    echo "${RED}模型下载失败。请手动从以下地址下载：${NC}"
-    echo "1. https://modelscope.cn/models/iic/SenseVoiceSmall/resolve/master/model.pt"
-    echo "2. 百度网盘: https://pan.baidu.com/share/init?surl=QlgM58FHhYv1tFnUT_A8Sg (提取码: qvna)"
-    echo "下载后请将文件放置在 models/SenseVoiceSmall/model.pt"
-fi
-
-# 下载配置文件
-echo "下载配置文件..."
-if [ "$DOWNLOAD_CMD" = "powershell -Command Invoke-WebRequest -Uri" ]; then
-    $DOWNLOAD_CMD "https://raw.githubusercontent.com/xinnan-tech/xiaozhi-esp32-server/main/main/xiaozhi-server/docker-compose.yml" $DOWNLOAD_CMD_SUFFIX "docker-compose.yml"
-    $DOWNLOAD_CMD "https://raw.githubusercontent.com/xinnan-tech/xiaozhi-esp32-server/main/main/xiaozhi-server/config.yaml" $DOWNLOAD_CMD_SUFFIX "data/.config.yaml"
-else
-    $DOWNLOAD_CMD "docker-compose.yml" "https://raw.githubusercontent.com/xinnan-tech/xiaozhi-esp32-server/main/main/xiaozhi-server/docker-compose.yml"
-    $DOWNLOAD_CMD "data/.config.yaml" "https://raw.githubusercontent.com/xinnan-tech/xiaozhi-esp32-server/main/main/xiaozhi-server/config.yaml"
-fi
-
-# 检查文件是否存在
-echo "检查文件完整性..."
-FILES_TO_CHECK="docker-compose.yml data/.config.yaml models/SenseVoiceSmall/model.pt"
-ALL_FILES_EXIST=true
-
-for FILE in $FILES_TO_CHECK; do
-    if [ ! -f "$FILE" ]; then
-        echo "${RED}错误: $FILE 不存在${NC}"
-        ALL_FILES_EXIST=false
-    fi
-done
-
-if [ "$ALL_FILES_EXIST" = false ]; then
-    echo "${RED}某些文件下载失败，请检查上述错误信息并手动下载缺失的文件。${NC}"
     exit 1
 fi
 
-echo "${GREEN}文件下载完成！${NC}"
+# 检查系统版本
-echo "请编辑 data/.config.yaml 文件配置你的API密钥。"
+if [ -f /etc/os-release ]; then
-echo "配置完成后，运行以下命令启动服务："
+    . /etc/os-release
-echo "${GREEN}docker-compose up -d${NC}"
+    if [ "$ID" != "debian" ] && [ "$ID" != "ubuntu" ]; then
-echo "查看日志请运行："
+        whiptail --title "系统错误" --msgbox "该脚本只支持Debian/Ubuntu系统执行" 10 60
-echo "${GREEN}docker logs -f xiaozhi-esp32-server${NC}"
+        exit 1
+    fi
+else
+    whiptail --title "系统错误" --msgbox "无法确定系统版本，该脚本只支持Debian/Ubuntu系统执行" 10 60
+    exit 1
+fi
 
-# 提示用户编辑配置文件
+# 下载配置文件函数
-echo "\n${RED}重要提示：${NC}"
+check_and_download() {
-echo "1. 请确保编辑 data/.config.yaml 文件，配置必要的API密钥"
+    local filepath=$1
-echo "2. 特别是 ChatGLM 和 mem0ai 的密钥必须配置"
+    local url=$2
-echo "3. 配置完成后再启动 docker 服务"
+    if [ ! -f "$filepath" ]; then
+        if ! curl -fL --progress-bar "$url" -o "$filepath"; then
+            whiptail --title "错误" --msgbox "${filepath}文件下载失败" 10 50
+            exit 1
+        fi
+    else
+        echo "${filepath}文件已存在，跳过下载"
+    fi
+}
+
+# 检查是否已安装
+check_installed() {
+    # 检查目录是否存在且非空
+    if [ -d "/opt/xiaozhi-server/" ] && [ "$(ls -A /opt/xiaozhi-server/)" ]; then
+        DIR_CHECK=1
+    else
+        DIR_CHECK=0
+    fi
+    
+    # 检查容器是否存在
+    if docker inspect xiaozhi-esp32-server > /dev/null 2>&1; then
+        CONTAINER_CHECK=1
+    else
+        CONTAINER_CHECK=0
+    fi
+    
+    # 两次检查都通过
+    if [ $DIR_CHECK -eq 1 ] && [ $CONTAINER_CHECK -eq 1 ]; then
+        return 0  # 已安装
+    else
+        return 1  # 未安装
+    fi
+}
+
+# 更新相关
+if check_installed; then
+    if whiptail --title "已安装检测" --yesno "检测到小智服务端已安装，是否进行升级？" 10 60; then
+        # 用户选择升级，执行清理操作
+        echo "开始升级操作..."
+        
+        # 停止并移除所有docker-compose服务
+        docker compose -f /opt/xiaozhi-server/docker-compose_all.yml down
+        
+        # 停止并删除特定容器（考虑容器可能不存在的情况）
+        containers=(
+            "xiaozhi-esp32-server"
+            "xiaozhi-esp32-server-web"
+            "xiaozhi-esp32-server-db"
+            "xiaozhi-esp32-server-redis"
+        )
+        
+        for container in "${containers[@]}"; do
+            if docker ps -a --format '{{.Names}}' | grep -q "^${container}$"; then
+                docker stop "$container" >/dev/null 2>&1 && \
+                docker rm "$container" >/dev/null 2>&1 && \
+                echo "成功移除容器: $container"
+            else
+                echo "容器不存在，跳过: $container"
+            fi
+        done
+        
+        # 删除特定镜像（考虑镜像可能不存在的情况）
+        images=(
+            "ghcr.nju.edu.cn/xinnan-tech/xiaozhi-esp32-server:server_latest"
+            "ghcr.nju.edu.cn/xinnan-tech/xiaozhi-esp32-server:web_latest"
+        )
+        
+        for image in "${images[@]}"; do
+            if docker images --format '{{.Repository}}:{{.Tag}}' | grep -q "^${image}$"; then
+                docker rmi "$image" >/dev/null 2>&1 && \
+                echo "成功删除镜像: $image"
+            else
+                echo "镜像不存在，跳过: $image"
+            fi
+        done
+        
+        echo "所有清理操作完成"
+        
+        # 备份原有配置文件
+        mkdir -p /opt/xiaozhi-server/backup/
+        if [ -f /opt/xiaozhi-server/data/.config.yaml ]; then
+            cp /opt/xiaozhi-server/data/.config.yaml /opt/xiaozhi-server/backup/.config.yaml
+            echo "已备份原有配置文件到 /opt/xiaozhi-server/backup/.config.yaml"
+        fi
+        
+        # 下载最新版配置文件
+        check_and_download "/opt/xiaozhi-server/docker-compose_all.yml" "https://ghfast.top/https://raw.githubusercontent.com/xinnan-tech/xiaozhi-esp32-server/refs/heads/main/main/xiaozhi-server/docker-compose_all.yml"
+        check_and_download "/opt/xiaozhi-server/data/.config.yaml" "https://ghfast.top/https://raw.githubusercontent.com/xinnan-tech/xiaozhi-esp32-server/refs/heads/main/main/xiaozhi-server/config_from_api.yaml"
+        
+        # 启动Docker服务
+        echo "开始启动最新版本服务..."
+        # 升级完成后标记，跳过后续下载步骤
+        UPGRADE_COMPLETED=1
+        docker compose -f /opt/xiaozhi-server/docker-compose_all.yml up -d
+    else
+          whiptail --title "跳过升级" --msgbox "已取消升级，将继续使用当前版本。" 10 50
+          # 跳过升级，继续执行后续安装流程
+    fi
+fi
+
+
+# 检查curl安装
+if ! command -v curl &> /dev/null; then
+    echo "------------------------------------------------------------"
+    echo "未检测到curl，正在安装..."
+    apt update
+    apt install -y curl
+else
+    echo "------------------------------------------------------------"
+    echo "curl已安装，跳过安装步骤"
+fi
+
+# 检查Docker安装
+if ! command -v docker &> /dev/null; then
+    echo "------------------------------------------------------------"
+    echo "未检测到Docker，正在安装..."
+    
+    # 使用国内镜像源替代官方源
+    DISTRO=$(lsb_release -cs)
+    MIRROR_URL="https://mirrors.aliyun.com/docker-ce/linux/ubuntu"
+    GPG_URL="https://mirrors.aliyun.com/docker-ce/linux/ubuntu/gpg"
+    
+    # 安装基础依赖
+    apt update
+    apt install -y apt-transport-https ca-certificates curl software-properties-common gnupg
+    
+    # 创建密钥目录并添加国内镜像源密钥
+    mkdir -p /etc/apt/keyrings
+    curl -fsSL "$GPG_URL" | gpg --dearmor -o /etc/apt/keyrings/docker.gpg
+    
+    # 添加国内镜像源
+    echo "deb [arch=$(dpkg --print-architecture) signed-by=/etc/apt/keyrings/docker.gpg] $MIRROR_URL $DISTRO stable" \
+        > /etc/apt/sources.list.d/docker.list
+    
+    # 添加备用官方源密钥（避免国内源密钥验证失败）
+    apt-key adv --keyserver keyserver.ubuntu.com --recv-keys 7EA0A9C3F273FCD8 2>/dev/null || \
+    echo "警告：部分密钥添加失败，继续尝试安装..."
+    
+    # 安装Docker
+    apt update
+    apt install -y docker-ce docker-ce-cli containerd.io
+    
+    # 启动服务
+    systemctl start docker
+    systemctl enable docker
+    
+    # 检查是否安装成功
+    if docker --version; then
+        echo "------------------------------------------------------------"
+        echo "Docker安装完成！"
+    else
+        whiptail --title "错误" --msgbox "Docker安装失败，请检查日志。" 10 50
+        exit 1
+    fi
+else
+    echo "Docker已安装，跳过安装步骤"
+fi
+
+# Docker镜像源配置
+MIRROR_OPTIONS=(
+    "1" "轩辕镜像 (推荐)"
+    "2" "腾讯云镜像源"
+    "3" "中科大镜像源"
+    "4" "网易163镜像源"
+    "5" "华为云镜像源"
+    "6" "阿里云镜像源"
+    "7" "自定义镜像源"
+    "8" "跳过配置"
+)
+
+MIRROR_CHOICE=$(whiptail --title "选择Docker镜像源" --menu "请选择要使用的Docker镜像源" 20 60 10 \
+"${MIRROR_OPTIONS[@]}" 3>&1 1>&2 2>&3) || {
+    echo "用户取消选择，退出脚本"
+    exit 1
+}
+
+case $MIRROR_CHOICE in
+    1) MIRROR_URL="https://docker.xuanyuan.me" ;; 
+    2) MIRROR_URL="https://mirror.ccs.tencentyun.com" ;; 
+    3) MIRROR_URL="https://docker.mirrors.ustc.edu.cn" ;; 
+    4) MIRROR_URL="https://hub-mirror.c.163.com" ;; 
+    5) MIRROR_URL="https://05f073ad3c0010ea0f4bc00b7105ec20.mirror.swr.myhuaweicloud.com" ;; 
+    6) MIRROR_URL="https://registry.aliyuncs.com" ;; 
+    7) MIRROR_URL=$(whiptail --title "自定义镜像源" --inputbox "请输入完整的镜像源URL:" 10 60 3>&1 1>&2 2>&3) ;; 
+    8) MIRROR_URL="" ;; 
+esac
+
+if [ -n "$MIRROR_URL" ]; then
+    mkdir -p /etc/docker
+    if [ -f /etc/docker/daemon.json ]; then
+        cp /etc/docker/daemon.json /etc/docker/daemon.json.bak
+    fi
+    cat > /etc/docker/daemon.json <<EOF
+{
+    "dns": ["8.8.8.8", "114.114.114.114"],
+    "registry-mirrors": ["$MIRROR_URL"]
+}
+EOF
+    whiptail --title "配置成功" --msgbox "已成功添加镜像源: $MIRROR_URL\n请按Enter键重启Docker服务并继续..." 12 60
+    echo "------------------------------------------------------------"
+    echo "开始重启Docker服务..."
+    systemctl restart docker.service
+fi
+
+# 创建安装目录
+echo "------------------------------------------------------------"
+echo "开始创建安装目录..."
+# 检查并创建数据目录
+if [ ! -d /opt/xiaozhi-server/data ]; then
+    mkdir -p /opt/xiaozhi-server/data
+    echo "已创建数据目录: /opt/xiaozhi-server/data"
+else
+    echo "目录xiaozhi-server/data已存在，跳过创建"
+fi
+
+# 检查并创建模型目录
+if [ ! -d /opt/xiaozhi-server/models/SenseVoiceSmall ]; then
+    mkdir -p /opt/xiaozhi-server/models/SenseVoiceSmall
+    echo "已创建模型目录: /opt/xiaozhi-server/models/SenseVoiceSmall"
+else
+    echo "目录xiaozhi-server/models/SenseVoiceSmall已存在，跳过创建"
+fi
+
+echo "------------------------------------------------------------"
+echo "开始下载语音识别模型"
+# 下载模型文件
+MODEL_PATH="/opt/xiaozhi-server/models/SenseVoiceSmall/model.pt"
+if [ ! -f "$MODEL_PATH" ]; then
+    (
+    for i in {1..20}; do
+        echo $((i*5))
+        sleep 0.5
+    done
+    ) | whiptail --title "下载中" --gauge "开始下载语音识别模型..." 10 60 0
+    curl -fL --progress-bar https://modelscope.cn/models/iic/SenseVoiceSmall/resolve/master/model.pt -o "$MODEL_PATH" || {
+        whiptail --title "错误" --msgbox "model.pt文件下载失败" 10 50
+        exit 1
+    }
+else
+    echo "model.pt文件已存在，跳过下载"
+fi
+
+# 如果不是升级完成，才执行下载
+if [ -z "$UPGRADE_COMPLETED" ]; then
+    check_and_download "/opt/xiaozhi-server/docker-compose_all.yml" "https://ghfast.top/https://raw.githubusercontent.com/xinnan-tech/xiaozhi-esp32-server/refs/heads/main/main/xiaozhi-server/docker-compose_all.yml"
+    check_and_download "/opt/xiaozhi-server/data/.config.yaml" "https://ghfast.top/https://raw.githubusercontent.com/xinnan-tech/xiaozhi-esp32-server/refs/heads/main/main/xiaozhi-server/config_from_api.yaml"
+fi
+
+# 启动Docker服务
+(
+echo "------------------------------------------------------------"
+echo "正在拉取Docker镜像..."
+echo "这可能需要几分钟时间，请耐心等待"
+docker compose -f /opt/xiaozhi-server/docker-compose_all.yml up -d
+
+if [ $? -ne 0 ]; then
+    whiptail --title "错误" --msgbox "Docker服务启动失败，请尝试更换镜像源后重新执行本脚本" 10 60
+    exit 1
+fi
+
+echo "------------------------------------------------------------"
+echo "正在检查服务启动状态..."
+TIMEOUT=300
+START_TIME=$(date +%s)
+while true; do
+    CURRENT_TIME=$(date +%s)
+    if [ $((CURRENT_TIME - START_TIME)) -gt $TIMEOUT ]; then
+        whiptail --title "错误" --msgbox "服务启动超时，未在指定时间内找到预期日志内容" 10 60
+        exit 1
+    fi
+    
+    if docker logs xiaozhi-esp32-server-web 2>&1 | grep -q "Started AdminApplication in"; then
+        break
+    fi
+    sleep 1
+done
+
+    echo "服务端启动成功！正在完成配置..."
+    echo "正在启动服务..."
+    docker compose -f docker-compose_all.yml up -d
+    echo "服务启动完成！"
+)
+
+# 密钥配置
+
+# 获取服务器公网地址
+PUBLIC_IP=$(hostname -I | awk '{print $1}')
+whiptail --title "配置服务器密钥" --msgbox "请使用浏览器，访问下方链接，打开智控台并注册账号: \n\n内网地址：http://127.0.0.1:8002/\n公网地址：http://$PUBLIC_IP:8002/ (若是云服务器请在服务器安全组放行端口 8000 8001 8002)。\n\n注册的第一个用户即是超级管理员，以后注册的用户都是普通用户。普通用户只能绑定设备和配置智能体; 超级管理员可以进行模型管理、用户管理、参数配置等功能。\n\n注册好后请按Enter键继续" 18 70
+SECRET_KEY=$(whiptail --title "配置服务器密钥" --inputbox "请使用超级管理员账号登录智控台\n内网地址：http://127.0.0.1:8002/\n公网地址：http://$PUBLIC_IP:8002/\n在顶部菜单 参数字典 → 参数管理 找到参数编码: server.secret (服务器密钥) \n复制该参数值并输入到下面输入框\n\n请输入密钥(留空则跳过配置):" 15 60 3>&1 1>&2 2>&3)
+
+if [ -n "$SECRET_KEY" ]; then
+    python3 -c "
+import sys, yaml; 
+config_path = '/opt/xiaozhi-server/data/.config.yaml'; 
+with open(config_path, 'r') as f: 
+    config = yaml.safe_load(f) or {}; 
+config['manager-api'] = {'url': 'http://xiaozhi-esp32-server-web:8002/xiaozhi', 'secret': '$SECRET_KEY'}; 
+with open(config_path, 'w') as f: 
+    yaml.dump(config, f); 
+"
+    docker restart xiaozhi-esp32-server
+fi
+
+# 获取并显示地址信息
+LOCAL_IP=$(hostname -I | awk '{print $1}')
+WEBSOCKET_ADDR=$(docker logs xiaozhi-esp32-server 2>&1 | tac | grep -m 1 -E -o "ws://[^ ]+")
+VISION_ADDR=$(docker logs xiaozhi-esp32-server 2>&1 | tac | grep -m 1 "视觉" | grep -m 1 -E -o "http://[^ ]+")
+
+whiptail --title "安装完成！" --msgbox "\
+服务端相关地址如下：\n\
+管理后台访问地址: http://$LOCAL_IP:8002\n\
+OTA 地址: http://$LOCAL_IP:8002/xiaozhi/ota/\n\
+视觉分析接口地址: $VISION_ADDR\n\
+WebSocket 地址: $WEBSOCKET_ADDR\n\
+\n安装完毕！感谢您的使用！\n按Enter键退出..." 16 70

docs/Deployment.md

@@ -8,45 +8,13 @@ docker镜像已支持x86架构、arm64架构的CPU，支持在国产操作系统
 
 如果您的电脑还没安装docker，可以按照这里的教程安装：[docker安装](https://www.runoob.com/docker/ubuntu-docker-install.html)
 
-如果你已经安装好docker，你可以[1.1使用懒人脚本](#11-懒人脚本)自动帮你下载所需的文件和配置文件，你可以使用docker[1.2手动部署](#12-手动部署)。
+安装好docker后，进继续。
 
-### 1.1 懒人脚本
+### 1.1 手动部署
-
-你可以使用以下命令一键下载并执行部署脚本：
-请确保你的环境可以正常访问 GitHub 否则无法下载脚本。
-```bash
-curl -L -o docker-setup.sh https://raw.githubusercontent.com/xinnan-tech/xiaozhi-esp32-server/main/docker-setup.sh
-```
-
-如果您的电脑是windows系统，请使用使用 Git Bash、WSL、PowerShell 或 CMD 运行以下命令：
-```bash
-# Git Bash 或 WSL
-sh docker-setup.sh
-# PowerShell 或 CMD
-.\docker-setup.sh
-```
-
-如果您的电脑是linux 或者 macos 系统，请使用终端运行以下命令：
-```bash
-chmod +x docker-setup.sh
-./docker-setup.sh
-```
-
-脚本会自动完成以下操作：
-> 1. 创建必要的目录结构
-> 2. 下载语音识别模型
-> 3. 下载配置文件
-> 4. 检查文件完整性
->
-> 执行完成后，请按照提示配置 API 密钥。
-
-当你一切顺利完成以上操作后，继续操作[配置项目文件](#2-配置项目文件)
-
-### 1.2 手动部署
 
 如果懒人脚本无法正常运行，请按本章节1.2进行手动部署。
 
-#### 1.2.1 创建目录
+#### 1.1.1 创建目录
 
 安装完后，你需要为这个项目找一个安放配置文件的目录，例如我们可以新建一个文件夹叫`xiaozhi-server`。
 
@@ -61,18 +29,18 @@ xiaozhi-server
      ├─ SenseVoiceSmall
 ```
 
-#### 1.2.2 下载语音识别模型文件
+#### 1.1.2 下载语音识别模型文件
 
 你需要下载语音识别的模型文件，因为本项目的默认语音识别用的是本地离线语音识别方案。可通过这个方式下载
 [跳转到下载语音识别模型文件](#模型文件)
 
 下载完后，回到本教程。
 
-#### 1.2.3 下载配置文件
+#### 1.1.3 下载配置文件
 
 你需要下载两个配置文件：`docker-compose.yaml` 和 `config.yaml`。需要从项目仓库下载这两个文件。
 
-##### 1.2.3.1 下载 docker-compose.yaml
+##### 1.1.3.1 下载 docker-compose.yaml
 
 用浏览器打开[这个链接](../main/xiaozhi-server/docker-compose.yml)。
 
@@ -81,7 +49,7 @@ xiaozhi-server
 
 下载完后，回到本教程继续往下。
 
-##### 1.2.3.2 创建 config.yaml
+##### 1.1.3.2 创建 config.yaml
 
 用浏览器打开[这个链接](../main/xiaozhi-server/config.yaml)。
 
@@ -260,7 +228,7 @@ LLM:
 文件放在`models/SenseVoiceSmall`
 目录下。下面两个下载路线任选一个。
 
-- 线路一：阿里魔塔下载[SenseVoiceSmall](https://modelscope.cn/models/iic/SenseVoiceSmall/resolve/master/model.pt)
+- 线路一：阿里魔搭下载[SenseVoiceSmall](https://modelscope.cn/models/iic/SenseVoiceSmall/resolve/master/model.pt)
 - 线路二：百度网盘下载[SenseVoiceSmall](https://pan.baidu.com/share/init?surl=QlgM58FHhYv1tFnUT_A8Sg&pwd=qvna) 提取码:
   `qvna`
 

docs/Deployment_all.md

@@ -7,7 +7,32 @@ docker镜像已支持x86架构、arm64架构的CPU，支持在国产操作系统
 
 如果您的电脑还没安装docker，可以按照这里的教程安装：[docker安装](https://www.runoob.com/docker/ubuntu-docker-install.html)
 
-#### 1.1 创建目录
+docker 安装全模块有两种方式，你可以[1.1使用懒人脚本](#1.1 懒人脚本)（作者[@VanillaNahida](https://github.com/VanillaNahida)）自动帮你下载所需的文件和配置文件，你可以使用[1.2手动部署](#1.2 手动部署)从零搭建。
+
+### 1.1 懒人脚本
+
+你可以使用以下命令一键安装全模块版小智服务端：  
+> [!NOTE]  
+> 暂且只支持Ubuntu服务器一键部署，其他系统未尝试，可能会有一些奇怪的bug
+
+使用SSH工具连接到服务器，以root权限执行如下脚本
+```bash
+sudo bash -c "$(wget -qO- https://ghfast.top/https://raw.githubusercontent.com/xinnan-tech/xiaozhi-esp32-server/main/docker-setup.sh)"
+```
+
+脚本会自动完成以下操作：
+> 1. 安装Docker
+> 2. 配置镜像源
+> 3. 下载/拉取镜像
+> 4. 下载语音识别模型文件
+> 5. 引导配置服务端
+>
+
+执行完成后简单配置后，再参照[4. 运行程序](#4. 运行程序)和[5.重启xiaozhi-esp32-server](#5.重启xiaozhi-esp32-server)里提到的最重要的3件事情，完成3这三项配置后即可使用。
+
+### 1.2 手动部署
+
+#### 1.2.1 创建目录
 
 安装完后，你需要为这个项目找一个安放配置文件的目录，例如我们可以新建一个文件夹叫`xiaozhi-server`。
 
@@ -22,22 +47,22 @@ xiaozhi-server
      ├─ SenseVoiceSmall
 ```
 
-#### 1.2 下载语音识别模型文件
+#### 1.2.2 下载语音识别模型文件
 
 本项目语音识别模型，默认使用`SenseVoiceSmall`模型，进行语音转文字。因为模型较大，需要独立下载，下载后把`model.pt`
 文件放在`models/SenseVoiceSmall`
 目录下。下面两个下载路线任选一个。
 
-- 线路一：阿里魔塔下载[SenseVoiceSmall](https://modelscope.cn/models/iic/SenseVoiceSmall/resolve/master/model.pt)
+- 线路一：阿里魔搭下载[SenseVoiceSmall](https://modelscope.cn/models/iic/SenseVoiceSmall/resolve/master/model.pt)
 - 线路二：百度网盘下载[SenseVoiceSmall](https://pan.baidu.com/share/init?surl=QlgM58FHhYv1tFnUT_A8Sg&pwd=qvna) 提取码:
   `qvna`
 
 
-#### 1.3 下载配置文件
+#### 1.2.3 下载配置文件
 
 你需要下载两个配置文件：`docker-compose_all.yaml` 和 `config_from_api.yaml`。需要从项目仓库下载这两个文件。
 
-##### 1.3.1 下载 docker-compose_all.yaml
+##### 1.2.3.1 下载 docker-compose_all.yaml
 
 用浏览器打开[这个链接](../main/xiaozhi-server/docker-compose_all.yml)。
 
@@ -48,7 +73,7 @@ xiaozhi-server
 
 下载完后，回到本教程继续往下。
 
-##### 1.3.2 下载 config_from_api.yaml
+##### 1.2.3.2 下载 config_from_api.yaml
 
 用浏览器打开[这个链接](../main/xiaozhi-server/config_from_api.yaml)。
 
@@ -179,12 +204,12 @@ docker logs -f xiaozhi-esp32-server
 
 OTA接口：
 ```
-http://你电脑局域网的ip:8002/xiaozhi/ota/
+http://你宿主机局域网的ip:8002/xiaozhi/ota/
 ```
 
 Websocket接口：
 ```
-ws://你电脑局域网的ip:8000/xiaozhi/v1/
+ws://你宿主机的ip:8000/xiaozhi/v1/
 ```
 
 ### 第三件重要的事情
@@ -358,7 +383,7 @@ pip install -r requirements.txt
 文件放在`models/SenseVoiceSmall`
 目录下。下面两个下载路线任选一个。
 
-- 线路一：阿里魔塔下载[SenseVoiceSmall](https://modelscope.cn/models/iic/SenseVoiceSmall/resolve/master/model.pt)
+- 线路一：阿里魔搭下载[SenseVoiceSmall](https://modelscope.cn/models/iic/SenseVoiceSmall/resolve/master/model.pt)
 - 线路二：百度网盘下载[SenseVoiceSmall](https://pan.baidu.com/share/init?surl=QlgM58FHhYv1tFnUT_A8Sg&pwd=qvna) 提取码:
   `qvna`
 

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

@@ -237,7 +237,7 @@ public interface Constant {
     /**
      * 版本号
      */
-    public static final String VERSION = "0.7.2";
+    public static final String VERSION = "0.7.3";
 
     /**
      * 无效固件URL

main/manager-api/src/main/java/xiaozhi/common/utils/JsonRpcTwo.java

@@ -0,0 +1,21 @@
+package xiaozhi.common.utils;
+
+import lombok.Data;
+/**
+ * JSON-RPC2.0 格式规范对象
+ */
+@Data
+public class JsonRpcTwo {
+    private String jsonrpc = "2.0";
+    private String method;
+    private Object params;
+    private Integer id;
+
+    public JsonRpcTwo(String method, Object params, Integer id) {
+        this.method = method;
+        this.params = params;
+        this.id = id;
+    }
+
+
+}

main/manager-api/src/main/java/xiaozhi/modules/agent/Enums/XiaoZhiMcpJsonRpcJson.java

@@ -0,0 +1,44 @@
+package xiaozhi.modules.agent.Enums;
+
+import xiaozhi.common.utils.JsonUtils;
+import xiaozhi.common.utils.JsonRpcTwo;
+
+import java.util.Map;
+
+
+/**
+ * 小智MCP JSON-RPC 请求json
+ */
+public class XiaoZhiMcpJsonRpcJson {
+    //小智初始化mcp请求json
+    private static final String INITIALIZE_JSON;
+    //小智mcp初始化成功，返回通知请求json
+    private static final String NOTIFICATIONS_INITIALIZED_JSON;
+    //小智mcp获取mcp工具集合请求json
+    private static final String TOOLS_LIST_REQUEST;
+    // 延迟加载
+    static {
+        INITIALIZE_JSON = JsonUtils.toJsonString(new JsonRpcTwo("initialize",
+                Map.of(
+                        "protocolVersion", "2024-11-05",
+                        "capabilities", Map.of(
+                                "roots", Map.of("listChanged", false),
+                                "sampling", Map.of()),
+                        "clientInfo", Map.of(
+                                "name", "xz-mcp-broker",
+                                "version", "0.0.1")),
+                1));
+        NOTIFICATIONS_INITIALIZED_JSON = "{\"jsonrpc\":\"2.0\",\"method\":\"notifications/initialized\"}";
+        TOOLS_LIST_REQUEST = JsonUtils.toJsonString(new JsonRpcTwo("tools/list", null, 2));
+    }
+    public static String getInitializeJson(){
+        return INITIALIZE_JSON;
+    }
+    public static String getNotificationsInitializedJson(){
+        return NOTIFICATIONS_INITIALIZED_JSON;
+    }
+    public static String getToolsListJson(){
+        return TOOLS_LIST_REQUEST;
+    }
+
+}

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

@@ -1,32 +0,0 @@
-package xiaozhi.modules.agent.dto;
-
-import lombok.Data;
-
-/**
- * MCP JSON-RPC 请求 DTO
- */
-@Data
-public class McpJsonRpcRequest {
-    private String jsonrpc = "2.0";
-    private String method;
-    private Object params;
-    private Integer id;
-
-    public McpJsonRpcRequest() {
-    }
-
-    public McpJsonRpcRequest(String method) {
-        this.method = method;
-    }
-
-    public McpJsonRpcRequest(String method, Object params, Integer id) {
-        this.method = method;
-        this.params = params;
-        this.id = id;
-    }
-
-    public McpJsonRpcRequest(String method, Object params) {
-        this.method = method;
-        this.params = params;
-    }
-}

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

@@ -103,7 +103,11 @@ public class AgentChatHistoryServiceImpl extends ServiceImpl<AiAgentChatHistoryD
         wrapper.select(AgentChatHistoryEntity::getContent, AgentChatHistoryEntity::getAudioId)
                 .eq(AgentChatHistoryEntity::getAgentId, agentId)
                 .eq(AgentChatHistoryEntity::getChatType, AgentChatHistoryType.USER.getValue())
-                .isNotNull(AgentChatHistoryEntity::getAudioId);
+                .isNotNull(AgentChatHistoryEntity::getAudioId)
+                // 添加此行，确保查询结果按照创建时间降序排列
+                // 使用id的原因：数据形式，id越大的创建时间就越晚，所以使用id的结果和创建时间降序排列结果一样
+                // id作为降序排列的优势，性能高，有主键索引，不用在排序的时候重新进行排除扫描比较
+                .orderByDesc(AgentChatHistoryEntity::getId); 
 
         // 构建分页查询，查询前50页数据
         Page<AgentChatHistoryEntity> pageParam = new Page<>(0, 50);

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

@@ -18,7 +18,7 @@ import xiaozhi.common.constant.Constant;
 import xiaozhi.common.utils.AESUtils;
 import xiaozhi.common.utils.HashEncryptionUtil;
 import xiaozhi.common.utils.JsonUtils;
-import xiaozhi.modules.agent.dto.McpJsonRpcRequest;
+import xiaozhi.modules.agent.Enums.XiaoZhiMcpJsonRpcJson;
 import xiaozhi.modules.agent.service.AgentMcpAccessPointService;
 import xiaozhi.modules.sys.service.SysParamsService;
 import xiaozhi.modules.sys.utils.WebSocketClientManager;
@@ -71,17 +71,7 @@ public class AgentMcpAccessPointServiceImpl implements AgentMcpAccessPointServic
 
                 // 步骤1: 发送初始化消息并等待响应
                 log.info("发送MCP初始化消息，智能体ID: {}", id);
-                McpJsonRpcRequest initializeRequest = new McpJsonRpcRequest("initialize",
+                client.sendText(XiaoZhiMcpJsonRpcJson.getInitializeJson());
-                        Map.of(
-                                "protocolVersion", "2024-11-05",
-                                "capabilities", Map.of(
-                                        "roots", Map.of("listChanged", false),
-                                        "sampling", Map.of()),
-                                "clientInfo", Map.of(
-                                        "name", "xz-mcp-broker",
-                                        "version", "0.0.1")),
-                        1);
-                client.sendJson(initializeRequest);
 
                 // 等待初始化响应 (id=1) - 移除固定延迟，改为响应驱动
                 List<String> initResponses = client.listenerWithoutClose(response -> {
@@ -125,13 +115,10 @@ public class AgentMcpAccessPointServiceImpl implements AgentMcpAccessPointServic
 
                 // 步骤2: 发送初始化完成通知 - 只有在收到initialize响应后才发送
                 log.info("发送MCP初始化完成通知，智能体ID: {}", id);
-                String notificationJson = "{\"jsonrpc\":\"2.0\",\"method\":\"notifications/initialized\"}";
+                client.sendText(XiaoZhiMcpJsonRpcJson.getNotificationsInitializedJson());
-                client.sendText(notificationJson);
-
                 // 步骤3: 发送工具列表请求 - 立即发送，无需额外延迟
                 log.info("发送MCP工具列表请求，智能体ID: {}", id);
-                McpJsonRpcRequest toolsRequest = new McpJsonRpcRequest("tools/list", null, 2);
+                client.sendText(XiaoZhiMcpJsonRpcJson.getToolsListJson());
-                client.sendJson(toolsRequest);
 
                 // 等待工具列表响应 (id=2)
                 List<String> toolsResponses = client.listener(response -> {

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

@@ -41,6 +41,7 @@ import xiaozhi.modules.agent.service.AgentTemplateService;
 import xiaozhi.modules.agent.vo.AgentInfoVO;
 import xiaozhi.modules.device.service.DeviceService;
 import xiaozhi.modules.model.dto.ModelProviderDTO;
+import xiaozhi.modules.model.entity.ModelConfigEntity;
 import xiaozhi.modules.model.service.ModelConfigService;
 import xiaozhi.modules.model.service.ModelProviderService;
 import xiaozhi.modules.security.user.SecurityUser;
@@ -324,9 +325,32 @@ public class AgentServiceImpl extends BaseServiceImpl<AgentDao, AgentEntity> imp
             // 删除音频数据
             agentChatHistoryService.deleteByAgentId(existingEntity.getId(), true, false);
         }
+
+        boolean b = validateLLMIntentParams(dto.getLlmModelId(), dto.getIntentModelId());
+        if (!b) {
+            throw new RenException("LLM大模型和Intent意图识别，选择参数不匹配");
+        }
         this.updateById(existingEntity);
     }
 
+    /**
+     * 验证大语言模型和意图识别的参数是否符合匹配
+     * 
+     * @param llmModelId    大语言模型id
+     * @param intentModelId 意图识别id
+     * @return T 匹配 : F 不匹配
+     */
+    private boolean validateLLMIntentParams(String llmModelId, String intentModelId) {
+        ModelConfigEntity llmModelData = modelConfigService.selectById(llmModelId);
+        String type = llmModelData.getConfigJson().get("type").toString();
+        // 如果查询大语言模型是openai或者ollama，意图识别选参数都可以
+        if ("openai".equals(type) || "ollama".equals(type)) {
+            return true;
+        }
+        // 除了openai和ollama的类型，不可以选择id为Intent_function_call（函数调用）的意图识别
+        return !"Intent_function_call".equals(intentModelId);
+    }
+
     @Override
     @Transactional(rollbackFor = Exception.class)
     public String createAgent(AgentCreateDTO dto) {

main/manager-api/src/main/java/xiaozhi/modules/model/controller/ModelController.java

@@ -21,11 +21,7 @@ import xiaozhi.common.utils.ConvertUtils;
 import xiaozhi.common.utils.Result;
 import xiaozhi.modules.agent.service.AgentTemplateService;
 import xiaozhi.modules.config.service.ConfigService;
-import xiaozhi.modules.model.dto.ModelBasicInfoDTO;
+import xiaozhi.modules.model.dto.*;
-import xiaozhi.modules.model.dto.ModelConfigBodyDTO;
-import xiaozhi.modules.model.dto.ModelConfigDTO;
-import xiaozhi.modules.model.dto.ModelProviderDTO;
-import xiaozhi.modules.model.dto.VoiceDTO;
 import xiaozhi.modules.model.entity.ModelConfigEntity;
 import xiaozhi.modules.model.service.ModelConfigService;
 import xiaozhi.modules.model.service.ModelProviderService;
@@ -52,6 +48,14 @@ public class ModelController {
         return new Result<List<ModelBasicInfoDTO>>().ok(modelList);
     }
 
+    @GetMapping("/llm/names")
+    @Operation(summary = "获取LLM模型信息")
+    @RequiresPermissions("sys:role:normal")
+    public Result<List<LlmModelBasicInfoDTO>> getLlmModelCodeList(@RequestParam(required = false) String modelName) {
+        List<LlmModelBasicInfoDTO> llmModelCodeList = modelConfigService.getLlmModelCodeList(modelName);
+        return new Result<List<LlmModelBasicInfoDTO>>().ok(llmModelCodeList);
+    }
+
     @GetMapping("/{modelType}/provideTypes")
     @Operation(summary = "获取模型供应器列表")
     @RequiresPermissions("sys:role:superAdmin")

main/manager-api/src/main/java/xiaozhi/modules/model/dto/LlmModelBasicInfoDTO.java

@@ -0,0 +1,13 @@
+package xiaozhi.modules.model.dto;
+
+import lombok.Data;
+import lombok.EqualsAndHashCode;
+
+/**
+ * LLM的模型的基础展示数据
+ */
+@EqualsAndHashCode(callSuper = true)
+@Data
+public class LlmModelBasicInfoDTO extends ModelBasicInfoDTO{
+    private String type;
+}

main/manager-api/src/main/java/xiaozhi/modules/model/service/ModelConfigService.java

@@ -4,6 +4,7 @@ import java.util.List;
 
 import xiaozhi.common.page.PageData;
 import xiaozhi.common.service.BaseService;
+import xiaozhi.modules.model.dto.LlmModelBasicInfoDTO;
 import xiaozhi.modules.model.dto.ModelBasicInfoDTO;
 import xiaozhi.modules.model.dto.ModelConfigBodyDTO;
 import xiaozhi.modules.model.dto.ModelConfigDTO;
@@ -13,6 +14,8 @@ public interface ModelConfigService extends BaseService<ModelConfigEntity> {
 
     List<ModelBasicInfoDTO> getModelCodeList(String modelType, String modelName);
 
+    List<LlmModelBasicInfoDTO> getLlmModelCodeList(String modelName);
+
     PageData<ModelConfigDTO> getPageList(String modelType, String modelName, String page, String limit);
 
     ModelConfigDTO add(String modelType, String provideCode, ModelConfigBodyDTO modelConfigBodyDTO);

main/manager-api/src/main/java/xiaozhi/modules/model/service/impl/ModelConfigServiceImpl.java

@@ -8,6 +8,7 @@ import java.util.stream.Collectors;
 import org.apache.commons.lang3.StringUtils;
 import org.springframework.stereotype.Service;
 
+import com.baomidou.mybatisplus.core.conditions.query.LambdaQueryWrapper;
 import com.baomidou.mybatisplus.core.conditions.query.QueryWrapper;
 import com.baomidou.mybatisplus.core.metadata.IPage;
 
@@ -23,6 +24,7 @@ import xiaozhi.common.utils.ConvertUtils;
 import xiaozhi.modules.agent.dao.AgentDao;
 import xiaozhi.modules.agent.entity.AgentEntity;
 import xiaozhi.modules.model.dao.ModelConfigDao;
+import xiaozhi.modules.model.dto.LlmModelBasicInfoDTO;
 import xiaozhi.modules.model.dto.ModelBasicInfoDTO;
 import xiaozhi.modules.model.dto.ModelConfigBodyDTO;
 import xiaozhi.modules.model.dto.ModelConfigDTO;
@@ -52,6 +54,25 @@ public class ModelConfigServiceImpl extends BaseServiceImpl<ModelConfigDao, Mode
         return ConvertUtils.sourceToTarget(entities, ModelBasicInfoDTO.class);
     }
 
+    @Override
+    public List<LlmModelBasicInfoDTO> getLlmModelCodeList(String modelName) {
+        List<ModelConfigEntity> entities = modelConfigDao.selectList(
+                new QueryWrapper<ModelConfigEntity>()
+                        .eq("model_type", "llm")
+                        .eq("is_enabled", 1)
+                        .like(StringUtils.isNotBlank(modelName), "model_name", "%" + modelName + "%")
+                        .select("id", "model_name", "config_json"));
+        // 处理获取到的内容
+        return entities.stream().map(item -> {
+            LlmModelBasicInfoDTO dto = new LlmModelBasicInfoDTO();
+            dto.setId(item.getId());
+            dto.setModelName(item.getModelName());
+            String type = item.getConfigJson().get("type").toString();
+            dto.setType(type);
+            return dto;
+        }).toList();
+    }
+
     @Override
     public PageData<ModelConfigDTO> getPageList(String modelType, String modelName, String page, String limit) {
         Map<String, Object> params = new HashMap<String, Object>();
@@ -94,6 +115,21 @@ public class ModelConfigServiceImpl extends BaseServiceImpl<ModelConfigDao, Mode
         if (CollectionUtil.isEmpty(providerList)) {
             throw new RenException("供应器不存在");
         }
+        if (modelConfigBodyDTO.getConfigJson().containsKey("llm")) {
+            String llm = modelConfigBodyDTO.getConfigJson().get("llm").toString();
+            ModelConfigEntity modelConfigEntity = modelConfigDao.selectOne(new LambdaQueryWrapper<ModelConfigEntity>()
+                    .eq(ModelConfigEntity::getId, llm));
+            String selectModelType = (modelConfigEntity == null || modelConfigEntity.getModelType() == null) ? null
+                    : modelConfigEntity.getModelType().toUpperCase();
+            if (modelConfigEntity == null || !"LLM".equals(selectModelType)) {
+                throw new RenException("设置的LLM不存在");
+            }
+            String type = modelConfigEntity.getConfigJson().get("type").toString();
+            // 如果查询大语言模型是openai或者ollama，意图识别选参数都可以
+            if (!"openai".equals(type) && !"ollama".equals(type)) {
+                throw new RenException("设置的LLM不是openai和ollama");
+            }
+        }
 
         // 再更新供应器提供的模型
         ModelConfigEntity modelConfigEntity = ConvertUtils.sourceToTarget(modelConfigBodyDTO, ModelConfigEntity.class);
@@ -137,6 +173,8 @@ public class ModelConfigServiceImpl extends BaseServiceImpl<ModelConfigDao, Mode
                         .or()
                         .eq("mem_model_id", modelId)
                         .or()
+                        .eq("vllm_model_id", modelId)
+                        .or()
                         .eq("intent_model_id", modelId));
         if (!agents.isEmpty()) {
             String agentNames = agents.stream()

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

@@ -106,6 +106,22 @@ export default {
         });
       }).send();
   },
+  // 获取LLM模型名称列表
+  getLlmModelCodeList(modelName, callback) {
+    RequestService.sendRequest()
+      .url(`${getServiceUrl()}/models/llm/names`)
+      .method('GET')
+      .data({ modelName })
+      .success((res) => {
+        RequestService.clearRequestTime();
+        callback(res);
+      })
+      .networkFail(() => {
+        RequestService.reAjaxFun(() => {
+          this.getLlmModelCodeList(modelName, callback);
+        });
+      }).send();
+  },
   // 获取模型音色列表
   getModelVoices(modelId, voiceName, callback) {
     const queryParams = new URLSearchParams({

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

@@ -160,10 +160,6 @@ export default {
     visible(newVal) {
       if (newVal) {
         this.dialogKey = Date.now();
-      }
-    },
-    agentId(newVal) {
-      if (newVal) {
         api.agent.getRecentlyFiftyByAgentId(this.agentId, ((data) => {
           this.valueTypeOptions = data.data.data.map(item => ({
             ...item

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

@@ -89,7 +89,7 @@
                       <div class="model-select-wrapper">
                         <el-select v-model="form.model[model.key]" filterable placeholder="请选择" class="form-select"
                           @change="handleModelChange(model.type, $event)">
-                          <el-option v-for="(item, optionIndex) in modelOptions[model.type]"
+                          <el-option v-for="(item, optionIndex) in modelOptions[model.type]" v-if="!item.isHidden"
                             :key="`option-${index}-${optionIndex}`" :label="item.label" :value="item.value" />
                         </el-select>
                         <div v-if="showFunctionIcons(model.type)" class="function-icons">
@@ -130,7 +130,6 @@
         </div>
       </div>
     </div>
-
     <function-dialog v-model="showFunctionDialog" :functions="currentFunctions" :all-functions="allFunctions"
       :agent-id="$route.query.agentId" @update-functions="handleUpdateFunctions" @dialog-closed="handleDialogClosed" />
   </div>
@@ -173,8 +172,9 @@ export default {
         { label: '视觉大模型(VLLM)', key: 'vllmModelId', type: 'VLLM' },
         { label: '意图识别(Intent)', key: 'intentModelId', type: 'Intent' },
         { label: '记忆(Memory)', key: 'memModelId', type: 'Memory' },
-        { label: '语音合成(TTS)', key: 'ttsModelId', type: 'TTS' },
+        { label: '语音合成(TTS)', key: 'ttsModelId', type: 'TTS' }
       ],
+      llmModeTypeMap: new Map(),
       modelOptions: {},
       templates: [],
       loadingTemplate: false,
@@ -356,6 +356,9 @@ export default {
             });
             // 备份原始，以备取消时恢复
             this.originalFunctions = JSON.parse(JSON.stringify(this.currentFunctions));
+
+            // 确保意图识别选项的可见性正确
+            this.updateIntentOptionsVisibility();
           });
         } else {
           this.$message.error(data.msg || '获取配置失败');
@@ -364,16 +367,41 @@ export default {
     },
     fetchModelOptions() {
       this.models.forEach(model => {
-        Api.model.getModelNames(model.type, '', ({ data }) => {
+        if (model.type != "LLM") {
-          if (data.code === 0) {
+          Api.model.getModelNames(model.type, '', ({ data }) => {
-            this.$set(this.modelOptions, model.type, data.data.map(item => ({
+            if (data.code === 0) {
-              value: item.id,
+              this.$set(this.modelOptions, model.type, data.data.map(item => ({
-              label: item.modelName
+                value: item.id,
-            })));
+                label: item.modelName,
-          } else {
+                isHidden: false
-            this.$message.error(data.msg || '获取模型列表失败');
+              })));
-          }
+
-        });
+              // 如果是意图识别选项，需要根据当前LLM类型更新可见性
+              if (model.type === 'Intent') {
+                this.updateIntentOptionsVisibility();
+              }
+            } else {
+              this.$message.error(data.msg || '获取模型列表失败');
+            }
+          });
+        } else {
+          Api.model.getLlmModelCodeList('', ({ data }) => {
+            if (data.code === 0) {
+              let LLMdata = []
+              data.data.forEach(item => {
+                LLMdata.push({
+                  value: item.id,
+                  label: item.modelName,
+                  isHidden: false
+                })
+                this.llmModeTypeMap.set(item.id, item.type)
+              })
+              this.$set(this.modelOptions, model.type, LLMdata);
+            } else {
+              this.$message.error(data.msg || '获取LLM模型列表失败');
+            }
+          });
+        }
       });
     },
     fetchVoiceOptions(modelId) {
@@ -410,6 +438,10 @@ export default {
       if (type === 'Memory' && value !== 'Memory_nomem' && (this.form.chatHistoryConf === 0 || this.form.chatHistoryConf === null)) {
         this.form.chatHistoryConf = 2;
       }
+      if (type === 'LLM') {
+        // 当LLM类型改变时，更新意图识别选项的可见性
+        this.updateIntentOptionsVisibility();
+      }
     },
     fetchAllFunctions() {
       return new Promise((resolve, reject) => {
@@ -450,6 +482,42 @@ export default {
       }
       this.showFunctionDialog = false;
     },
+    updateIntentOptionsVisibility() {
+      // 根据当前选择的LLM类型更新意图识别选项的可见性
+      const currentLlmId = this.form.model.llmModelId;
+      if (!currentLlmId || !this.modelOptions['Intent']) return;
+
+      const llmType = this.llmModeTypeMap.get(currentLlmId);
+      if (!llmType) return;
+
+      this.modelOptions['Intent'].forEach(item => {
+        if (item.value === "Intent_function_call") {
+          // 如果llmType是openai或ollama，允许选择function_call
+          // 否则隐藏function_call选项
+          if (llmType === "openai" || llmType === "ollama") {
+            item.isHidden = false;
+          } else {
+            item.isHidden = true;
+          }
+        } else {
+          // 其他意图识别选项始终可见
+          item.isHidden = false;
+        }
+      });
+
+      // 如果当前选择的意图识别是function_call，但LLM类型不支持，则设置为可选的第一项
+      if (this.form.model.intentModelId === "Intent_function_call" &&
+        llmType !== "openai" && llmType !== "ollama") {
+        // 找到第一个可见的选项
+        const firstVisibleOption = this.modelOptions['Intent'].find(item => !item.isHidden);
+        if (firstVisibleOption) {
+          this.form.model.intentModelId = firstVisibleOption.value;
+        } else {
+          // 如果没有可见选项，设置为Intent_nointent
+          this.form.model.intentModelId = 'Intent_nointent';
+        }
+      }
+    },
     updateChatHistoryConf() {
       if (this.form.model.memModelId === 'Memory_nomem') {
         this.form.chatHistoryConf = 0;

main/xiaozhi-server/config.yaml

@@ -271,9 +271,19 @@ ASR:
     api_key: none
     output_dir: tmp/
   SherpaASR:
+    # Sherpa-ONNX 本地语音识别（需手动下载模型）
     type: sherpa_onnx_local
     model_dir: models/sherpa-onnx-sense-voice-zh-en-ja-ko-yue-2024-07-17
     output_dir: tmp/
+    # 模型类型：sense_voice (多语言) 或 paraformer (中文专用)
+    model_type: sense_voice
+  SherpaParaformerASR:
+    # 中文语音识别模型，可以运行在低性能设备（需手动下载模型，例如RK3566-2g）
+    # 详细配置说明请参考：docs/sherpa-paraformer-guide.md
+    type: sherpa_onnx_local
+    model_dir: models/sherpa-onnx-paraformer-zh-small-2024-03-09
+    output_dir: tmp/
+    model_type: paraformer
   DoubaoASR:
     # 可以在这里申请相关Key等信息
     # https://console.volcengine.com/speech/app
@@ -710,6 +720,26 @@ TTS:
     #     voice_id: female-shaonv
     #     weight: 1
     # language_boost: auto
+
+# MinimaxTTSHTTPStream和MinimaxTTSWebSocketStream还在测试，测试完再开放
+#
+#  MinimaxTTSHTTPStream:
+#  # Minimax流式语音合成服务
+#    type: minimax_httpstream
+#    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"
+#    voice_id: "female-shaonv"
+#
+#  MinimaxTTSWebSocketStream:
+#    type: minimax_webSocket
+#    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"
+#    voice_id: "female-shaonv"
+
   AliyunTTS:
     # 阿里云智能语音交互服务，需要先在阿里云平台开通服务，然后获取验证信息
     # 平台地址：https://nls-portal.console.aliyun.com/

main/xiaozhi-server/config/logger.py

@@ -5,7 +5,7 @@ from config.config_loader import load_config
 from config.settings import check_config_file
 from datetime import datetime
 
-SERVER_VERSION = "0.7.2"
+SERVER_VERSION = "0.7.3"
 _logger_initialized = False
 
 

main/xiaozhi-server/core/connection.py

@@ -132,6 +132,8 @@ class ConnectionHandler:
 
         # tts相关变量
         self.sentence_id = None
+        # 处理TTS响应没有文本返回
+        self.tts_MessageText = ""
 
         # iot相关变量
         self.iot_descriptors = {}
@@ -182,8 +184,13 @@ class ConnectionHandler:
                     await ws.send("端口正常，如需测试连接，请使用test_page.html")
                     await self.close(ws)
                     return
-            # 获取客户端ip地址
+            real_ip = self.headers.get("x-real-ip") or self.headers.get(
-            self.client_ip = ws.remote_address[0]
+                "x-forwarded-for"
+            )
+            if real_ip:
+                self.client_ip = real_ip.split(",")[0].strip()
+            else:
+                self.client_ip = ws.remote_address[0]
             self.logger.bind(tag=TAG).info(
                 f"{self.client_ip} conn - Headers: {self.headers}"
             )
@@ -272,7 +279,6 @@ class ConnectionHandler:
     async def _route_message(self, message):
         """消息路由"""
         if isinstance(message, str):
-            self.last_activity_time = time.time() * 1000
             await handleTextMessage(self, message)
         elif isinstance(message, bytes):
             if self.vad is None:
@@ -335,7 +341,7 @@ class ConnectionHandler:
                 self.config.get("selected_module", {})
             )
             self.logger = create_connection_logger(self.selected_module_str)
-            
+
             """初始化组件"""
             if self.config.get("prompt") is not None:
                 user_prompt = self.config["prompt"]
@@ -351,10 +357,10 @@ class ConnectionHandler:
                 self.vad = self._vad
             if self.asr is None:
                 self.asr = self._initialize_asr()
-            
+
             # 初始化声纹识别
             self._initialize_voiceprint()
-            
+
             # 打开语音识别通道
             asyncio.run_coroutine_threadsafe(
                 self.asr.open_audio_channels(self), self.loop
@@ -746,7 +752,7 @@ class ConnectionHandler:
                 content = response
 
             # 在llm回复中获取情绪表情，一轮对话只在开头获取一次
-            if emotion_flag:
+            if emotion_flag and content is not None and content.strip():
                 asyncio.run_coroutine_threadsafe(
                     textUtils.get_emotion(self, content),
                     self.loop,
@@ -790,9 +796,9 @@ class ConnectionHandler:
             if not bHasError:
                 # 如需要大模型先处理一轮，添加相关处理后的日志情况
                 if len(response_message) > 0:
-                    self.dialogue.put(
+                    text_buff = "".join(response_message)
-                        Message(role="assistant", content="".join(response_message))
+                    self.tts_MessageText = text_buff
-                    )
+                    self.dialogue.put(Message(role="assistant", content=text_buff))
                 response_message.clear()
                 self.logger.bind(tag=TAG).debug(
                     f"function_name={function_name}, function_id={function_id}, function_arguments={function_arguments}"
@@ -814,9 +820,9 @@ class ConnectionHandler:
 
         # 存储对话内容
         if len(response_message) > 0:
-            self.dialogue.put(
+            text_buff = "".join(response_message)
-                Message(role="assistant", content="".join(response_message))
+            self.tts_MessageText = text_buff
-            )
+            self.dialogue.put(Message(role="assistant", content=text_buff))
         if depth == 0:
             self.tts.tts_text_queue.put(
                 TTSMessageDTO(
@@ -853,7 +859,7 @@ class ConnectionHandler:
                             {
                                 "id": function_id,
                                 "function": {
-                                    "arguments": function_arguments,
+                                    "arguments": "{}" if function_arguments == "" else function_arguments,
                                     "name": function_name,
                                 },
                                 "type": "function",
@@ -893,9 +899,7 @@ class ConnectionHandler:
                     if self.executor is None:
                         continue
                     # 提交任务到线程池
-                    self.executor.submit(
+                    self.executor.submit(self._process_report, *item)
-                        self._process_report, *item
-                    )
                 except Exception as e:
                     self.logger.bind(tag=TAG).error(f"聊天记录上报线程异常: {e}")
             except queue.Empty:

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

@@ -12,7 +12,7 @@ async def sendAudioMessage(conn, sentenceType, audios, text):
     conn.logger.bind(tag=TAG).info(f"发送音频消息: {sentenceType}, {text}")
 
     pre_buffer = False
-    if conn.tts.tts_audio_first_sentence and text is not None:
+    if conn.tts.tts_audio_first_sentence:
         conn.logger.bind(tag=TAG).info(f"发送第一段语音: {text}")
         conn.tts.tts_audio_first_sentence = False
         pre_buffer = True
@@ -21,8 +21,6 @@ async def sendAudioMessage(conn, sentenceType, audios, text):
 
     await sendAudio(conn, audios, pre_buffer)
 
-    await send_tts_message(conn, "sentence_end", text)
-
     # 发送结束消息（如果是最后一个文本）
     if conn.llm_finish_task and sentenceType == SentenceType.LAST:
         await send_tts_message(conn, "stop", None)
@@ -73,7 +71,7 @@ async def send_tts_message(conn, state, text=None):
     """发送 TTS 状态消息"""
     message = {"type": "tts", "state": state, "session_id": conn.session_id}
     if text is not None:
-        message["text"] = text
+        message["text"] = textUtils.check_emoji(text)
 
     # TTS播放结束
     if state == "stop":

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

@@ -1,4 +1,5 @@
 import json
+import time
 from core.handle.abortHandle import handleAbortMessage
 from core.handle.helloHandle import handleHelloMessage
 from core.providers.tools.device_mcp import handle_mcp_message
@@ -45,6 +46,7 @@ async def handleTextMessage(conn, message):
                 conn.client_have_voice = False
                 conn.asr_audio.clear()
                 if "text" in msg_json:
+                    conn.last_activity_time = time.time() * 1000
                     original_text = msg_json["text"]  # 保留原始文本
                     filtered_len, filtered_text = remove_punctuation_and_length(
                         original_text

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

@@ -84,7 +84,13 @@ class ASRProvider(ASRProviderBase):
         self.appkey = config.get("appkey")
         self.token = config.get("token")
         self.host = config.get("host", "nls-gateway-cn-shanghai.aliyuncs.com")
-        self.ws_url = f"wss://{self.host}/ws/v1"
+        # 如果配置的是内网地址（包含-internal.aliyuncs.com），则使用ws协议，默认是wss协议
+        if "-internal." in self.host:
+            self.ws_url = f"ws://{self.host}/ws/v1"
+        else:
+            # 默认使用wss协议
+            self.ws_url = f"wss://{self.host}/ws/v1"
+
         self.max_sentence_silence = config.get("max_sentence_silence")
         self.output_dir = config.get("output_dir", "./audio_output")
         self.delete_audio_file = delete_audio_file

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

@@ -40,6 +40,7 @@ class ASRProvider(ASRProviderBase):
         self.interface_type = InterfaceType.LOCAL
         self.model_dir = config.get("model_dir")
         self.output_dir = config.get("output_dir")
+        self.model_type = config.get("model_type", "sense_voice")  # 支持 paraformer
         self.delete_audio_file = delete_audio_file
 
         # 确保输出目录存在
@@ -73,16 +74,27 @@ class ASRProvider(ASRProviderBase):
             raise
 
         with CaptureOutput():
-            self.model = sherpa_onnx.OfflineRecognizer.from_sense_voice(
+            if self.model_type == "paraformer":
-                model=self.model_path,
+                self.model = sherpa_onnx.OfflineRecognizer.from_paraformer(
-                tokens=self.tokens_path,
+                    paraformer=self.model_path,
-                num_threads=2,
+                    tokens=self.tokens_path,
-                sample_rate=16000,
+                    num_threads=2,
-                feature_dim=80,
+                    sample_rate=16000,
-                decoding_method="greedy_search",
+                    feature_dim=80,
-                debug=False,
+                    decoding_method="greedy_search",
-                use_itn=True,
+                    debug=False,
-            )
+                )
+            else:  # sense_voice
+                self.model = sherpa_onnx.OfflineRecognizer.from_sense_voice(
+                    model=self.model_path,
+                    tokens=self.tokens_path,
+                    num_threads=2,
+                    sample_rate=16000,
+                    feature_dim=80,
+                    decoding_method="greedy_search",
+                    debug=False,
+                    use_itn=True,
+                )
 
     def read_wave(self, wave_filename: str) -> Tuple[np.ndarray, int]:
         """

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

@@ -120,15 +120,18 @@ class TTSProvider(TTSProviderBase):
 
         # WebSocket配置
         self.host = config.get("host", "nls-gateway-cn-beijing.aliyuncs.com")
-        self.ws_url = f"wss://{self.host}/ws/v1"
+        # 如果配置的是内网地址（包含-internal.aliyuncs.com），则使用ws协议，默认是wss协议
+        if "-internal." in self.host:
+            self.ws_url = f"ws://{self.host}/ws/v1"
+        else:
+            # 默认使用wss协议
+            self.ws_url = f"wss://{self.host}/ws/v1"
         self.ws = None
         self._monitor_task = None
         self.last_active_time = None
 
         # 专属tts设置
         self.message_id = ""
-        self.tts_text = ""
-        self.text_buffer = []
 
         # 创建Opus编码器
         self.opus_encoder = opus_encoder_utils.OpusEncoderUtils(
@@ -229,7 +232,6 @@ class TTSProvider(TTSProviderBase):
 
                         # aliyunStream独有的参数生成
                         self.message_id = str(uuid.uuid4().hex)
-                        self.text_buffer = []
 
                         logger.bind(tag=TAG).info("开始启动TTS会话...")
                         future = asyncio.run_coroutine_threadsafe(
@@ -250,7 +252,6 @@ class TTSProvider(TTSProviderBase):
                             logger.bind(tag=TAG).debug(
                                 f"开始发送TTS文本: {message.content_detail}"
                             )
-                            self.text_buffer.append(message.content_detail)
                             future = asyncio.run_coroutine_threadsafe(
                                 self.text_to_speak(message.content_detail, None),
                                 loop=self.conn.loop,
@@ -275,9 +276,6 @@ class TTSProvider(TTSProviderBase):
                 if message.sentence_type == SentenceType.LAST:
                     try:
                         logger.bind(tag=TAG).info("开始结束TTS会话...")
-                        self.tts_text = textUtils.get_string_no_punctuation_or_emoji(
-                            "".join(self.text_buffer).replace("\n", "")
-                        )
                         future = asyncio.run_coroutine_threadsafe(
                             self.finish_session(self.conn.sentence_id),
                             loop=self.conn.loop,
@@ -444,34 +442,35 @@ class TTSProvider(TTSProviderBase):
                             event_name = header.get("name")
                             if event_name == "SynthesisStarted":
                                 logger.bind(tag=TAG).debug("TTS合成已启动")
-                            elif event_name == "SentenceBegin":
-                                logger.bind(tag=TAG).debug(
-                                    f"句子语音生成开始: {self.tts_text}"
-                                )
-                                opus_datas_cache = []
                                 self.tts_audio_queue.put(
-                                    (SentenceType.FIRST, [], self.tts_text)
+                                    (SentenceType.FIRST, [], None)
                                 )
+                            elif event_name == "SentenceBegin":
+                                opus_datas_cache = []
                             elif event_name == "SentenceEnd":
-                                logger.bind(tag=TAG).info(
-                                    f"句子语音生成成功： {self.tts_text}"
-                                )
                                 if (
                                     not is_first_sentence
                                     or first_sentence_segment_count > 10
                                 ):
                                     # 发送缓存的数据
-                                    self.tts_audio_queue.put(
+                                    if self.conn.tts_MessageText:
-                                        (SentenceType.MIDDLE, opus_datas_cache, None)
+                                        logger.bind(tag=TAG).info(
-                                    )
+                                            f"句子语音生成成功： {self.conn.tts_MessageText}"
+                                        )
+                                        self.tts_audio_queue.put(
+                                            (SentenceType.MIDDLE, opus_datas_cache, self.conn.tts_MessageText)
+                                        )
+                                        self.conn.tts_MessageText = None
+                                    else:
+                                        self.tts_audio_queue.put(
+                                            (SentenceType.MIDDLE, opus_datas_cache, None)
+                                        )
                                 # 第一句话结束后，将标志设置为False
                                 is_first_sentence = False
                             elif event_name == "SynthesisCompleted":
                                 logger.bind(tag=TAG).debug(f"会话结束～～")
                                 self._process_before_stop_play_files()
                                 session_finished = True
-                                self.reuse_judgment = time.time()
-                                self.tts_text = ""
                                 break
                         except json.JSONDecodeError:
                             logger.bind(tag=TAG).warning("收到无效的JSON消息")

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

@@ -232,7 +232,6 @@ class TTSProvider(TTSProviderBase):
                             loop=self.conn.loop,
                         )
                         future.result()
-                        self.tts_audio_first_sentence = True
                         self.before_stop_play_files.clear()
                         logger.bind(tag=TAG).info("TTS会话启动成功")
                     except Exception as e:

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

@@ -52,7 +52,6 @@ class TTSProvider(TTSProviderBase):
                     self.processed_chars = 0
                     self.tts_text_buff = []
                     self.segment_count = 0
-                    self.tts_audio_first_sentence = True
                     self.before_stop_play_files.clear()
                 elif ContentType.TEXT == message.content_type:
                     self.tts_text_buff.append(message.content_detail)

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

@@ -0,0 +1,230 @@
+import os
+import uuid
+import json
+import requests
+from datetime import datetime
+from typing import Iterator, Optional, Union
+from core.providers.tts.base import TTSProviderBase
+from core.utils.util import parse_string_to_list
+
+
+class TTSProvider(TTSProviderBase):
+    def __init__(self, config, delete_audio_file):
+        super().__init__(config, delete_audio_file)
+        self.group_id = config.get("group_id")
+        self.api_key = config.get("api_key")
+        self.model = config.get("model")
+        if config.get("private_voice"):
+            self.voice = config.get("private_voice")
+        else:
+            self.voice = config.get("voice_id")
+
+        default_voice_setting = {
+            "voice_id": "female-shaonv",
+            "speed": 1,
+            "vol": 1,
+            "pitch": 0,
+            "emotion": "happy",
+        }
+        default_pronunciation_dict = {"tone": ["处理/(chu3)(li3)", "危险/dangerous"]}
+        defult_audio_setting = {
+            "sample_rate": 32000,
+            "bitrate": 128000,
+            "format": "mp3",
+            "channel": 1,
+        }
+        self.voice_setting = {
+            **default_voice_setting,
+            **config.get("voice_setting", {}),
+        }
+        self.pronunciation_dict = {
+            **default_pronunciation_dict,
+            **config.get("pronunciation_dict", {}),
+        }
+        self.audio_setting = {**defult_audio_setting, **config.get("audio_setting", {})}
+        self.timber_weights = parse_string_to_list(config.get("timber_weights"))
+
+        if self.voice:
+            self.voice_setting["voice_id"] = self.voice
+
+        self.host = "api.minimax.chat"
+        self.api_url = f"https://{self.host}/v1/t2a_v2?GroupId={self.group_id}"
+        self.header = {
+            "Content-Type": "application/json",
+            "Authorization": f"Bearer {self.api_key}",
+        }
+        self.audio_file_type = defult_audio_setting.get("format", "mp3")
+
+    def generate_filename(self, extension=".mp3"):
+        return os.path.join(
+            self.output_file,
+            f"tts-{__name__}{datetime.now().date()}@{uuid.uuid4().hex}{extension}",
+        )
+
+    async def text_to_speak(self, text, output_file):
+        """非流式语音合成（保留原有实现）"""
+        request_json = {
+            "model": self.model,
+            "text": text,
+            "stream": False,
+            "voice_setting": self.voice_setting,
+            "pronunciation_dict": self.pronunciation_dict,
+            "audio_setting": self.audio_setting,
+        }
+
+        if type(self.timber_weights) is list and len(self.timber_weights) > 0:
+            request_json["timber_weights"] = self.timber_weights
+            request_json["voice_setting"]["voice_id"] = ""
+
+        try:
+            resp = requests.post(
+                self.api_url, json.dumps(request_json), headers=self.header
+            )
+            if resp.json()["base_resp"]["status_code"] == 0:
+                data = resp.json()["data"]["audio"]
+                audio_bytes = bytes.fromhex(data)
+                if output_file:
+                    with open(output_file, "wb") as file_to_save:
+                        file_to_save.write(audio_bytes)
+                else:
+                    return audio_bytes
+            else:
+                raise Exception(
+                    f"{__name__} status_code: {resp.status_code} response: {resp.content}"
+                )
+        except Exception as e:
+            raise Exception(f"{__name__} error: {e}")
+
+    def text_to_speak_stream(
+        self, 
+        text: str, 
+        chunk_callback: Optional[callable] = None
+    ) -> Iterator[bytes]:
+        """
+        流式语音合成方法
+        :param text: 要合成的文本
+        :param chunk_callback: 可选的回调函数，用于处理每个音频块
+        :return: 生成器，每次产生一个音频数据块(bytes)
+        """
+        request_json = {
+            "model": self.model,
+            "text": text,
+            "stream": True,
+            "voice_setting": self.voice_setting,
+            "pronunciation_dict": self.pronunciation_dict,
+            "audio_setting": self.audio_setting,
+        }
+
+        if isinstance(self.timber_weights, list) and len(self.timber_weights) > 0:
+            request_json["timber_weights"] = self.timber_weights
+            request_json["voice_setting"]["voice_id"] = ""
+
+        try:
+            with requests.post(
+                self.api_url,
+                data=json.dumps(request_json),
+                headers=self.header,
+                stream=True
+            ) as response:
+                
+                # 检查HTTP状态码
+                if response.status_code != 200:
+                    raise Exception(
+                        f"HTTP error: {response.status_code}, response: {response.text}"
+                    )
+                
+                # 处理流式响应
+                for line in response.iter_lines():
+                    if line:  # 过滤空行
+                        # 检查是否为数据行 (SSE格式)
+                        if line.startswith(b'data:'):
+                            try:
+                                data = json.loads(line[5:].strip())  # 去掉"data:"前缀
+                                
+                                # 检查API状态码
+                                if data.get("base_resp", {}).get("status_code", -1) != 0:
+                                    raise Exception(
+                                        f"API error: {data.get('base_resp', {}).get('status_msg')}"
+                                    )
+                                
+                                # 跳过非音频数据块
+                                if "extra_info" in data:
+                                    continue
+                                    
+                                # 提取音频数据
+                                audio_hex = data.get("data", {}).get("audio")
+                                if audio_hex:
+                                    audio_chunk = bytes.fromhex(audio_hex)
+                                    if chunk_callback:
+                                        chunk_callback(audio_chunk)
+                                    yield audio_chunk
+                                    
+                            except json.JSONDecodeError:
+                                # 忽略JSON解析错误（可能是心跳包等）
+                                continue
+                            except Exception as e:
+                                raise e
+                                
+        except Exception as e:
+            raise Exception(f"{__name__} stream error: {e}")
+
+    def save_stream_to_file(
+        self, 
+        text: str, 
+        output_file: Optional[str] = None,
+        progress_callback: Optional[callable] = None
+    ) -> str:
+        """
+        流式合成并保存到文件
+        :param text: 要合成的文本
+        :param output_file: 输出文件路径，如果为None则自动生成
+        :param progress_callback: 可选的回调函数，接收已写入的字节数
+        :return: 保存的文件路径
+        """
+        if not output_file:
+            output_file = self.generate_filename(extension=f".{self.audio_file_type}")
+        
+        os.makedirs(os.path.dirname(output_file), exist_ok=True)
+        
+        total_bytes = 0
+        try:
+            with open(output_file, "wb") as audio_file:
+                for audio_chunk in self.text_to_speak_stream(text):
+                    audio_file.write(audio_chunk)
+                    audio_file.flush()
+                    total_bytes += len(audio_chunk)
+                    if progress_callback:
+                        progress_callback(total_bytes)
+            return output_file
+        except Exception as e:
+            # 清理可能创建的不完整文件
+            if os.path.exists(output_file):
+                os.remove(output_file)
+            raise e
+
+    def stream_to_audio_player(self, text: str, player_command: list = None):
+        """
+        流式合成并直接播放音频
+        :param text: 要合成的文本
+        :param player_command: 音频播放器命令，默认使用mpv
+        """
+        if player_command is None:
+            player_command = ["mpv", "--no-cache", "--no-terminal", "--", "fd://0"]
+        
+        try:
+            import subprocess
+            player_process = subprocess.Popen(
+                player_command,
+                stdin=subprocess.PIPE,
+                stdout=subprocess.DEVNULL,
+                stderr=subprocess.DEVNULL,
+            )
+            
+            for audio_chunk in self.text_to_speak_stream(text):
+                player_process.stdin.write(audio_chunk)
+                player_process.stdin.flush()
+                
+            player_process.stdin.close()
+            player_process.wait()
+        except Exception as e:
+            raise Exception(f"Audio player error: {e}")

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

@@ -0,0 +1,180 @@
+import os
+import uuid
+import json
+import asyncio
+import websockets
+import ssl
+from datetime import datetime
+from core.providers.tts.base import TTSProviderBase
+from core.utils.util import parse_string_to_list
+
+
+class TTSProvider(TTSProviderBase):
+    def __init__(self, config, delete_audio_file):
+        super().__init__(config, delete_audio_file)
+        self.group_id = config.get("group_id")
+        self.api_key = config.get("api_key")
+        self.model = config.get("model")
+        
+        # 初始化语音设置
+        default_voice_setting = {
+            "voice_id": "female-shaonv",
+            "speed": 1,
+            "vol": 1,
+            "pitch": 0,
+            "emotion": "happy",
+        }
+        default_pronunciation_dict = {"tone": ["处理/(chu3)(li3)", "危险/dangerous"]}
+        default_audio_setting = {
+            "sample_rate": 32000,
+            "bitrate": 128000,
+            "format": "mp3",
+            "channel": 1,
+        }
+        
+        # 合并配置
+        self.voice_setting = {
+            **default_voice_setting,
+            **config.get("voice_setting", {}),
+        }
+        self.pronunciation_dict = {
+            **default_pronunciation_dict,
+            **config.get("pronunciation_dict", {}),
+        }
+        self.audio_setting = {
+            **default_audio_setting, 
+            **config.get("audio_setting", {})
+        }
+        self.timber_weights = parse_string_to_list(config.get("timber_weights"))
+        
+        # 设置语音ID
+        if config.get("private_voice"):
+            self.voice_setting["voice_id"] = config.get("private_voice")
+        elif config.get("voice_id"):
+            self.voice_setting["voice_id"] = config.get("voice_id")
+        
+        # WebSocket配置
+        self.ws_url = "wss://api.minimaxi.com/ws/v1/t2a_v2"
+        self.headers = {
+            "Authorization": f"Bearer {self.api_key}",
+            "GroupId": self.group_id
+        }
+        self.audio_file_type = self.audio_setting.get("format", "mp3")
+
+    def generate_filename(self, extension=".mp3"):
+        """生成唯一的音频文件名"""
+        return os.path.join(
+            self.output_file,
+            f"tts-{__name__}{datetime.now().date()}@{uuid.uuid4().hex}{extension}",
+        )
+
+    async def _establish_connection(self):
+        """建立WebSocket连接"""
+        ssl_context = ssl.create_default_context()
+        ssl_context.check_hostname = False
+        ssl_context.verify_mode = ssl.CERT_NONE
+
+        try:
+            ws = await websockets.connect(
+                self.ws_url,
+                additional_headers=self.headers,
+                ssl=ssl_context
+            )
+            connected = json.loads(await ws.recv())
+            if connected.get("event") == "connected_success":
+                print("连接成功")
+                return ws
+            return None
+        except Exception as e:
+            print(f"连接失败: {e}")
+            return None
+
+    async def _start_task(self, websocket):
+        """发送任务开始请求"""
+        start_msg = {
+            "event": "task_start",
+            "model": self.model,
+            "voice_setting": self.voice_setting,
+            "pronunciation_dict": self.pronunciation_dict,
+            "audio_setting": self.audio_setting
+        }
+        
+        if self.timber_weights and len(self.timber_weights) > 0:
+            start_msg["timber_weights"] = self.timber_weights
+            start_msg["voice_setting"]["voice_id"] = ""
+
+        await websocket.send(json.dumps(start_msg))
+        response = json.loads(await websocket.recv())
+        return response.get("event") == "task_started"
+
+    async def _continue_task(self, websocket, text):
+        """发送继续请求并收集音频数据"""
+        await websocket.send(json.dumps({
+            "event": "task_continue",
+            "text": text
+        }))
+
+        audio_chunks = []
+        while True:
+            response = json.loads(await websocket.recv())
+            if "data" in response and "audio" in response["data"]:
+                audio_chunks.append(response["data"]["audio"])
+            if response.get("is_final"):
+                break
+        return "".join(audio_chunks)
+
+    async def _close_connection(self, websocket):
+        """关闭连接"""
+        if websocket:
+            await websocket.send(json.dumps({"event": "task_finish"}))
+            await websocket.close()
+            print("连接已关闭")
+
+    async def text_to_speak(self, text, output_file=None):
+        """主方法：文本转语音"""
+        ws = await self._establish_connection()
+        if not ws:
+            raise Exception("无法建立WebSocket连接")
+
+        try:
+            if not await self._start_task(ws):
+                raise Exception("任务启动失败")
+
+            hex_audio = await self._continue_task(ws, text)
+            audio_bytes = bytes.fromhex(hex_audio)
+
+            # 保存到文件或返回二进制数据
+            if output_file:
+                with open(output_file, "wb") as f:
+                    f.write(audio_bytes)
+                print(f"音频已保存为{output_file}")
+                return output_file
+            else:
+                # 返回音频二进制数据（不播放）
+                return audio_bytes
+
+        finally:
+            await self._close_connection(ws)
+
+
+async def main():
+    """测试用主函数"""
+    # 示例配置
+    config = {
+        "group_id": "YOUR_GROUP_ID",  # 替换为实际的group_id
+        "api_key": "YOUR_API_KEY",    # 替换为实际的api_key
+        "model": "your-model",        # 替换为实际的模型名称
+        "voice_id": "male-qn-qingse",
+        "voice_setting": {
+            "speed": 1.2,
+            "emotion": "happy"
+        }
+    }
+    
+    tts = TTSProvider(config, delete_audio_file=True)
+    output_file = tts.generate_filename()
+    await tts.text_to_speak("这是一个测试文本，用于验证流式语音合成功能", output_file)
+
+
+if __name__ == "__main__":
+    asyncio.run(main())

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

@@ -73,6 +73,10 @@ class Dialogue:
         if system_message:
             # 基础系统提示
             enhanced_system_prompt = system_message.content
+            # 替换时间占位符
+            enhanced_system_prompt = enhanced_system_prompt.replace(
+                "{{current_time}}", datetime.now().strftime("%H:%M")
+            )
 
             # 添加说话人个性化描述
             try:

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

@@ -120,7 +120,6 @@ class PromptManager:
         from datetime import datetime
 
         now = datetime.now()
-        current_time = now.strftime("%H:%M")
         today_date = now.strftime("%Y-%m-%d")
         today_weekday = WEEKDAY_MAP[now.strftime("%A")]
         today_lunar = cnlunar.Lunar(now, godType="8char")
@@ -130,7 +129,7 @@ class PromptManager:
             today_lunar.lunarDayCn,
         )
 
-        return current_time, today_date, today_weekday, lunar_date
+        return today_date, today_weekday, lunar_date
 
     def _get_location_info(self, client_ip: str) -> str:
         """获取位置信息"""
@@ -199,7 +198,7 @@ class PromptManager:
 
         try:
             # 获取最新的时间信息（不缓存）
-            current_time, today_date, today_weekday, lunar_date = (
+            today_date, today_weekday, lunar_date = (
                 self._get_current_time_info()
             )
 
@@ -224,7 +223,7 @@ class PromptManager:
             template = Template(self.base_prompt_template)
             enhanced_prompt = template.render(
                 base_prompt=user_prompt,
-                current_time=current_time,
+                current_time="{{current_time}}",
                 today_date=today_date,
                 today_weekday=today_weekday,
                 lunar_date=lunar_date,

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

@@ -24,6 +24,15 @@ EMOJI_MAP = {
     "😘": "kissy",
     "😏": "confident",
 }
+EMOJI_RANGES = [
+    (0x1F600, 0x1F64F),
+    (0x1F300, 0x1F5FF),
+    (0x1F680, 0x1F6FF),
+    (0x1F900, 0x1F9FF),
+    (0x1FA70, 0x1FAFF),
+    (0x2600, 0x26FF),
+    (0x2700, 0x27BF),
+]
 
 
 def get_string_no_punctuation_or_emoji(s):
@@ -65,18 +74,7 @@ def is_punctuation_or_emoji(char):
     }
     if char.isspace() or char in punctuation_set:
         return True
-    # 检查表情符号（保留原有逻辑）
+    return is_emoji(char)
-    code_point = ord(char)
-    emoji_ranges = [
-        (0x1F600, 0x1F64F),
-        (0x1F300, 0x1F5FF),
-        (0x1F680, 0x1F6FF),
-        (0x1F900, 0x1F9FF),
-        (0x1FA70, 0x1FAFF),
-        (0x2600, 0x26FF),
-        (0x2700, 0x27BF),
-    ]
-    return any(start <= code_point <= end for start, end in emoji_ranges)
 
 
 async def get_emotion(conn, text):
@@ -102,3 +100,14 @@ async def get_emotion(conn, text):
     except Exception as e:
         conn.logger.bind(tag=TAG).warning(f"发送情绪表情失败，错误:{e}")
     return
+
+
+def is_emoji(char):
+    """检查字符是否为emoji表情"""
+    code_point = ord(char)
+    return any(start <= code_point <= end for start, end in EMOJI_RANGES)
+
+
+def check_emoji(text):
+    """去除文本中的所有emoji表情"""
+    return ''.join(char for char in text if not is_emoji(char) and char != "\n")

main/xiaozhi-server/plugins_func/functions/hass_set_state.py

@@ -50,7 +50,9 @@ hass_set_state_function_desc = {
 
 
 @register_function("hass_set_state", hass_set_state_function_desc, ToolType.SYSTEM_CTL)
-def hass_set_state(conn, entity_id="", state={}):
+def hass_set_state(conn, entity_id="", state=None):
+    if state is None:
+        state = {}
     try:
         future = asyncio.run_coroutine_threadsafe(
             handle_hass_set_state(conn, entity_id, state), conn.loop
@@ -157,7 +159,7 @@ async def handle_hass_set_state(conn, entity_id, state):
         if domain == "vacuum":
             action = "start"
     else:
-        return f"{domain} {state.type}功能尚未支持"
+        return f"{domain} {state['type']}功能尚未支持"
 
     if arg == "":
         data = {