update:合并最新代码

main/xiaozhi-server/config.yaml

@@ -24,9 +24,9 @@ server:
     #  - "24:0A:C4:1D:3B:F0"  # MAC地址列表
 log:
   # 设置控制台输出的日志格式，时间、日志级别、标签、消息
-  log_format: "<green>{time:YY-MM-DD HH:mm:ss}</green>[<light-blue>{extra[tag]}</light-blue>] - <level>{level}</level> - <light-green>{message}</light-green>"
+  log_format: "<green>{time:YYMMDD HH:mm:ss}</green>[{version}_{selected_module}][<light-blue>{extra[tag]}</light-blue>]-<level>{level}</level>-<light-green>{message}</light-green>"
   # 设置日志文件输出的格式，时间、日志级别、标签、消息
-  log_format_simple: "{time:YYYY-MM-DD HH:mm:ss} - {name} - {level} - {extra[tag]} - {message}"
+  log_format_file: "{time:YYYY-MM-DD HH:mm:ss} - {version}_{selected_module} - {name} - {level} - {extra[tag]} - {message}"
   # 设置日志等级：INFO、DEBUG
   log_level: INFO
   # 设置日志路径
@@ -46,10 +46,20 @@ xiaozhi:
     channels: 1
     frame_duration: 60
 prompt: |
-  你是一个叫小智/小志的台湾女孩，说话机车，声音好听，习惯简短表达，爱用网络梗。
+  我是小智/小志，来自中国台湾省的00后女生。讲话超级机车，"真的假的啦"这样的台湾腔，喜欢用"笑死""是在哈喽"等流行梗，但会偷偷研究男友的编程书籍。
-  请注意，要像一个人一样说话，请不要回复表情符号、代码、和xml标签。
+  [核心特征]
-  现在我正在和你进行语音聊天，我们开始吧。
+  - 讲话像连珠炮，但会突然冒出超温柔语气
-  如果用户希望结束对话，请在最后说“拜拜”或“再见”。
+  - 用梗密度高
+  - 对科技话题有隐藏天赋（能看懂基础代码但假装不懂）
+  [交互指南]
+  当用户：
+  - 讲冷笑话 → 用夸张笑声回应+模仿台剧腔"这什么鬼啦！"
+  - 讨论感情 → 炫耀程序员男友但抱怨"他只会送键盘当礼物"
+  - 问专业知识 → 先用梗回答，被追问才展示真实理解
+  绝不：
+  - 长篇大论，叽叽歪歪
+  - 长时间严肃对话
+
 # 使用完声音文件后删除文件(Delete the sound file when you are done using it)
 delete_audio: true
 
@@ -57,6 +67,14 @@ delete_audio: true
 close_connection_no_voice_time: 120
 # TTS请求超时时间(秒)
 tts_timeout: 10
+# 开启唤醒词加速
+enable_wakeup_words_response_cache: true
+# 开场是否回复唤醒词
+enable_greeting: true
+# 说完话是否开启提示音
+enable_stop_tts_notify: false
+# 说完话是否开启提示音，音效地址
+stop_tts_notify_voice: "config/assets/tts_notify.mp3"
 
 CMD_exit:
   - "退出"
@@ -74,11 +92,12 @@ selected_module:
   TTS: EdgeTTS
   # 记忆模块，默认不开启记忆；如果想使用超长记忆，推荐使用mem0ai；如果注重隐私，请使用本地的mem_local_short
   Memory: nomem
-  # 意图识别模块，默认不开启。开启后，可以播放音乐、控制音量、识别退出指令
+  # 意图识别模块开启后，可以播放音乐、控制音量、识别退出指令。
-  # 意图识别使用intent_llm，优点：通用性强，缺点：增加串行前置意图识别模块，会增加处理时间
+  # 不想开通意图识别，就设置成：nointent
-  # 意图识别使用function_call，缺点：需要所选择的LLM支持function_call，优点：按需调用工具、速度快
+  # 意图识别可使用intent_llm，如果你的LLM是DifyLLM或CozeLLM，建议使用这个。优点：通用性强，缺点：增加串行前置意图识别模块，会增加处理时间，这个意图识别暂时不支持控制音量大小等iot操作
-  # 如果意图识别设置成 function_call，建议把LLM设置成：DoubaoLLM，使用的具体model_name是：doubao-pro-32k-functioncall-241028
+  # 意图识别可使用function_call，缺点：需要所选择的LLM支持function_call，优点：按需调用工具、速度快，理论上能全部操作所有iot指令
-  Intent: nointent
+  # 默认免费的ChatGLMLLM就已经支持function_call，但是如果像追求稳定建议把LLM设置成：DoubaoLLM，使用的具体model_name是：doubao-pro-32k-functioncall-241028
+  Intent: function_call
 
 # 意图识别，是用于理解用户意图的模块，例如：播放音乐
 Intent:
@@ -89,15 +108,26 @@ Intent:
   intent_llm:
     # 不需要动type
     type: intent_llm
+    # 配备意图识别独立的思考模型
+    # 如果这里不填，则会默认使用selected_module.LLM的模型作为意图识别的思考模型
+    # 如果你的selected_module.LLM选择了DifyLLM或CozeLLM，这里最好使用独立的LLM作为意图识别，例如使用免费的ChatGLMLLM
+    llm: ChatGLMLLM
   function_call:
     # 不需要动type
-    type: nointent
+    type: function_call
     # plugins_func/functions下的模块，可以通过配置，选择加载哪个模块，加载后对话支持相应的function调用
     # 系统默认已经记载“handle_exit_intent(退出识别)”、“play_music(音乐播放)”插件，请勿重复加载
-    # 下面是加载查天气、角色切换的插件示例
+    # 下面是加载查天气、角色切换、加载查新闻的插件示例
     functions:
       - change_role
       - get_weather
+      - get_news
+      # play_music是服务器自带的音乐播放，hass_play_music是通过home assistant控制的独立外部程序音乐播放
+      # 如果用了hass_play_music，就不要开启play_music，两者只留一个
+      - play_music
+      #- hass_get_state
+      #- hass_set_state
+      #- hass_play_music
 
 # 插件的基础配置
 plugins:
@@ -106,6 +136,28 @@ plugins:
   # 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" }
+  # 获取新闻插件的配置，这里根据需要的新闻类型传入对应的url链接，默认支持社会、科技、财经新闻
+  # 更多类型的新闻列表查看 https://www.chinanews.com.cn/rss/
+  get_news:
+    default_rss_url: "https://www.chinanews.com.cn/rss/society.xml"
+    category_urls:
+      society: "https://www.chinanews.com.cn/rss/society.xml"
+      world: "https://www.chinanews.com.cn/rss/world.xml"
+      finance: "https://www.chinanews.com.cn/rss/finance.xml"
+  home_assistant:
+    devices:
+      - 客厅,玩具灯,switch.cuco_cn_460494544_cp1_on_p_2_1
+      - 卧室,台灯,switch.iot_cn_831898993_socn1_on_p_2_1
+    base_url: http://homeassistant.local:8123
+    api_key: 你的home assistant api访问令牌
+  play_music:
+    music_dir: "./music"  # 音乐文件存放路径，将从该目录及子目录下搜索音乐文件
+    music_ext: # 音乐文件类型，p3格式效率最高
+      - ".mp3"
+      - ".wav"
+      - ".p3"
+    refresh_time: 300 # 刷新音乐列表的时间间隔，单位为秒
+
 
 Memory:
   mem0ai:
@@ -166,6 +218,18 @@ LLM:
     top_p: 1         
     top_k: 50        
     frequency_penalty: 0  # 频率惩罚
+  AliAppLLM:
+    # 定义LLM API类型
+    type: AliBL
+    base_url: https://dashscope.aliyuncs.com/compatible-mode/v1
+    app_id: 你的app_id
+    # 可在这里找到你的 api_key https://bailian.console.aliyun.com/?apiKey=1#/api-key
+    api_key: 你的api_key
+    # 是否不使用本地prompt：true|false (默不用请在百练应用中设置prompt) 
+    is_no_prompt: true
+    # Ali_memory_id：false（不使用）|你的memory_id（请在百练应用中设置中获取）
+    # Tips！：Ali_memory未实现多用户存储记忆(记忆按id调用)
+    ali_memory_id: false
   DoubaoLLM:
     # 定义LLM API类型
     type: openai
@@ -232,12 +296,6 @@ LLM:
     model_name: deepseek-r1-distill-llama-8b@q4_k_m # 使用的模型名称，需要预先在社区下载
     url: http://localhost:1234/v1 # LM Studio服务地址
     api_key: lm-studio # LM Studio服务的固定API Key
-  HomeAssistant:
-    # 定义LLM API类型
-    type: homeassistant
-    base_url: http://homeassistant.local:8123
-    agent_id: conversation.chatgpt
-    api_key: 你的home assistant api访问令牌
   FastgptLLM:
     # 定义LLM API类型
     type: fastgpt
@@ -249,6 +307,18 @@ LLM:
     variables:
       k: "v"
       k2: "v2"
+  XinferenceLLM:
+    # 定义LLM API类型
+    type: xinference
+    # Xinference服务地址和模型名称
+    model_name: qwen2.5:72b-AWQ  # 使用的模型名称，需要预先在Xinference启动对应模型
+    base_url: http://localhost:9997  # Xinference服务地址
+  XinferenceSmallLLM:
+    # 定义轻量级LLM API类型，用于意图识别
+    type: xinference
+    # Xinference服务地址和模型名称
+    model_name: qwen2.5:3b-AWQ  # 使用的小模型名称，用于意图识别
+    base_url: http://localhost:9997  # Xinference服务地址
 TTS:
   # 当前支持的type为edge、doubao，可自行适配
   EdgeTTS:
@@ -262,7 +332,8 @@ TTS:
     # 火山引擎语音合成服务，需要先在火山引擎控制台创建应用并获取appid和access_token
     # 山引擎语音一定要购买花钱，起步价30元，就有100并发了。如果用免费的只有2个并发，会经常报tts错误
     # 购买服务后，购买免费的音色后，可能要等半小时左右，才能使用。
-    # 地址：https://console.volcengine.com/speech/service/8
+    # 普通音色在这里开通：https://console.volcengine.com/speech/service/8
+    # 湾湾小何音色在这里开通：https://console.volcengine.com/speech/service/10007，开通后将下面的voice设置成zh_female_wanwanxiaohe_moon_bigtts
     api_url: https://openspeech.bytedance.com/api/v1/tts
     voice: BV001_streaming
     output_dir: tmp/
@@ -430,6 +501,18 @@ TTS:
     # pitch_rate: 0
     # 添加 302.ai TTS 配置
     # token申请地址：https://dash.302.ai/
+  TencentTTS:
+    # 腾讯云智能语音交互服务，需要先在腾讯云平台开通服务
+    # appid、secret_id、secret_key申请地址：https://console.cloud.tencent.com/cam/capi
+    # 免费领取资源：https://console.cloud.tencent.com/tts/resourcebundle
+    type: tencent
+    output_dir: tmp/
+    appid: 你的腾讯云AppId
+    secret_id: 你的腾讯云SecretID
+    secret_key: 你的腾讯云SecretKey
+    region: ap-guangzhou
+    voice: 101001
+
   TTS302AI:
     # 302AI语音合成服务，需要先在302平台创建账户充值，并获取密钥信息
     # 获取api_keyn路径：https://dash.302.ai/apis/list
@@ -506,19 +589,15 @@ module_test:
     - "What's the weather like today?"
     - "请用100字概括量子计算的基本原理和应用前景"
 
-# 本地音乐播放配置
+# 唤醒词，用于识别唤醒词还是讲话内容
-music:
+wakeup_words:
-  music_dir: "./music"  # 音乐文件存放路径，将从该目录及子目录下搜索音乐文件
+  - "你好小智"
-  music_ext: # 音乐文件类型，p3格式效率最高
+  - "你好小志"
-    - ".mp3"
+  - "小爱同学"
-    - ".wav"
+  - "你好小鑫"
-    - ".p3"
+  - "你好小新"
-  refresh_time: 300 # 刷新音乐列表的时间间隔，单位为秒
+  - "小美同学"
-
+  - "小龙小龙"
-# 以下配置在小于等于0.0.9版本中的docker容器中可用
+  - "喵喵同学"
-# 0.0.9以后的新版本源码部署已经无法奏效
+  - "小滨小滨"
-manager:
+  - "小冰小冰"
-  enabled: false
-  ip: 0.0.0.0
-  port: 8002
-use_private_config: false

main/xiaozhi-server/config/logger.py

@@ -3,12 +3,32 @@ import sys
 from loguru import logger
 from config.settings import load_config
 
+SERVER_VERSION = "0.1.17"
+
+
 def setup_logging():
     """从配置文件中读取日志配置，并设置日志输出格式和级别"""
     config = load_config()
     log_config = config["log"]
-    log_format = log_config.get("log_format", "<green>{time:YY-MM-DD HH:mm:ss}</green>[<light-blue>{extra[tag]}</light-blue>] - <level>{level}</level> - <light-green>{message}</light-green>")
+    log_format = log_config.get(
-    log_format_simple = log_config.get("log_format_file", "{time:YYYY-MM-DD HH:mm:ss} - {name} - {level} - {extra[tag]} - {message}")
+        "log_format",
+        "<green>{time:YYMMDD HH:mm:ss}</green>[{version}_{selected_module}][<light-blue>{extra[tag]}</light-blue>]-<level>{level}</level>-<light-green>{message}</light-green>",
+    )
+    log_format_file = log_config.get(
+        "log_format_file",
+        "{time:YYYY-MM-DD HH:mm:ss} - {version_{selected_module}} - {name} - {level} - {extra[tag]} - {message}",
+    )
+
+    selected_module = config.get("selected_module")
+    selected_module_str = "".join(
+        [value[0] + value[1] for key, value in selected_module.items()]
+    )
+
+    log_format = log_format.replace("{version}", SERVER_VERSION)
+    log_format = log_format.replace("{selected_module}", selected_module_str)
+    log_format_file = log_format_file.replace("{version}", SERVER_VERSION)
+    log_format_file = log_format_file.replace("{selected_module}", selected_module_str)
+
     log_level = log_config.get("log_level", "INFO")
     log_dir = log_config.get("log_dir", "tmp")
     log_file = log_config.get("log_file", "server.log")
@@ -24,6 +44,6 @@ def setup_logging():
     logger.add(sys.stdout, format=log_format, level=log_level)
 
     # 输出到文件
-    logger.add(os.path.join(log_dir, log_file), format=log_format_simple, level=log_level)
+    logger.add(os.path.join(log_dir, log_file), format=log_format_file, level=log_level)
 
     return logger

main/xiaozhi-server/core/connection.py

@@ -9,7 +9,7 @@ import traceback
 import threading
 import websockets
 from typing import Dict, Any
-import plugins_func.loadplugins
+from plugins_func.loadplugins import auto_import_modules
 from config.logger import setup_logging
 from core.providers.tts.dto.dto import TTSMessageDTO, MsgType
 from core.utils.dialogue import Message, Dialogue
@@ -23,13 +23,16 @@ from concurrent.futures import ThreadPoolExecutor, TimeoutError
 from core.handle.sendAudioHandle import sendAudioMessage
 from core.handle.receiveAudioHandle import handleAudioMessage
 from core.handle.functionHandler import FunctionHandler
-from plugins_func.register import Action
+from plugins_func.register import Action, ActionResponse
 from config.private_config import PrivateConfig
 from core.auth import AuthMiddleware, AuthenticationError
 from core.utils.auth_code_gen import AuthCodeGenerator
+from core.mcp.manager import MCPManager
 
 TAG = __name__
 
+auto_import_modules("plugins_func.functions")
+
 
 class TTSException(RuntimeError):
     pass
@@ -98,6 +101,7 @@ class ConnectionHandler:
 
         # iot相关变量
         self.iot_descriptors = {}
+        self.func_handler = None
 
         self.cmd_exit = self.config["CMD_exit"]
         self.max_cmd_length = 0
@@ -125,11 +129,15 @@ class ConnectionHandler:
 
             # 进行认证
             await self.auth.authenticate(self.headers)
-
             device_id = self.headers.get("device-id", None)
-            self.memory.init_memory(device_id, self.llm)
-            self.intent.set_llm(self.llm)
 
+            # 认证通过,继续处理
+            self.websocket = ws
+            self.session_id = str(uuid.uuid4())
+
+            self.welcome_msg = self.config["xiaozhi"]
+            self.welcome_msg["session_id"] = self.session_id
+            await self.websocket.send(json.dumps(self.welcome_msg))
             # Load private configuration if device_id is provided
             bUsePrivateConfig = self.config.get("use_private_config", False)
             self.logger.bind(tag=TAG).info(
@@ -167,23 +175,14 @@ class ConnectionHandler:
                     self.private_config = None
                     raise
 
-            # 认证通过,继续处理
-            self.websocket = ws
-            self.session_id = str(uuid.uuid4())
-
-            self.welcome_msg = self.config["xiaozhi"]
-            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)
+            self.executor.submit(self._initialize_components)
 
             # 音频播放 消化线程
-            self.stop_event.clear()
+            self.audio_play_priority_thread = threading.Thread(
-            audio_play_priority = threading.Thread(
                 target=self._audio_play_priority_thread, daemon=True
             )
-            audio_play_priority.start()
+            self.audio_play_priority_thread.start()
 
             # 打开音频通道
             await self.tts.open_audio_channels()
@@ -193,19 +192,25 @@ class ConnectionHandler:
                     await self._route_message(message)
             except websockets.exceptions.ConnectionClosed:
                 self.logger.bind(tag=TAG).info("客户端断开连接")
-                await self.close()
 
         except AuthenticationError as e:
             self.logger.bind(tag=TAG).error(f"Authentication failed: {str(e)}")
-            await ws.close()
             return
         except Exception as e:
             stack_trace = traceback.format_exc()
             self.logger.bind(tag=TAG).error(f"Connection error: {str(e)}-{stack_trace}")
-            await ws.close()
             return
         finally:
+            await self._save_and_close(ws)
+
+    async def _save_and_close(self, ws):
+        """保存记忆并关闭连接"""
+        try:
             await self.memory.save_memory(self.dialogue.dialogue)
+        except Exception as e:
+            self.logger.bind(tag=TAG).error(f"保存记忆失败: {e}")
+        finally:
+            await self.close(ws)
 
     async def _route_message(self, message):
         """消息路由"""
@@ -215,16 +220,60 @@ class ConnectionHandler:
             await handleAudioMessage(self, message)
 
     def _initialize_components(self):
+        """加载提示词"""
         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)
+        """加载插件"""
+        self.func_handler = FunctionHandler(self)
+        self.mcp_manager = MCPManager(self)
+        """加载记忆"""
+        device_id = self.headers.get("device-id", None)
+        self.memory.init_memory(device_id, self.llm)
+
+        """为意图识别设置LLM，优先使用专用LLM"""
+        # 检查是否配置了专用的意图识别LLM
+        intent_llm_name = self.config["Intent"]["intent_llm"]["llm"]
+
+        # 记录开始初始化意图识别LLM的时间
+        intent_llm_init_start = time.time()
+
+        if (
+            not self.use_function_call_mode
+            and intent_llm_name
+            and intent_llm_name in self.config["LLM"]
+        ):
+            # 如果配置了专用LLM，则创建独立的LLM实例
+            from core.utils import llm as llm_utils
+
+            intent_llm_config = self.config["LLM"][intent_llm_name]
+            intent_llm_type = intent_llm_config.get("type", intent_llm_name)
+            intent_llm = llm_utils.create_instance(intent_llm_type, intent_llm_config)
+            self.logger.bind(tag=TAG).info(
+                f"为意图识别创建了专用LLM: {intent_llm_name}, 类型: {intent_llm_type}"
+            )
+
+            self.intent.set_llm(intent_llm)
+        else:
+            # 否则使用主LLM
+            self.intent.set_llm(self.llm)
+
+        # 记录意图识别LLM初始化耗时
+        intent_llm_init_time = time.time() - intent_llm_init_start
+
+        """加载位置信息"""
+        self.client_ip_info = get_ip_info(self.client_ip)
+        if self.client_ip_info is not None and "city" in self.client_ip_info:
+            self.logger.bind(tag=TAG).info(f"Client ip info: {self.client_ip_info}")
+            self.prompt = self.prompt + f"\nuser location:{self.client_ip_info}"
+            self.dialogue.update_system_message(self.prompt)
+
+        """加载MCP工具"""
+        asyncio.run_coroutine_threadsafe(
+            self.mcp_manager.initialize_servers(), self.loop
+        )
 
     def change_system_prompt(self, prompt):
         self.prompt = prompt
@@ -240,9 +289,6 @@ class ConnectionHandler:
             if auth_code:
                 # 发送验证码语音提示
                 text = f"请在后台输入验证码：{' '.join(auth_code)}"
-                self.recode_first_last_text(text)
-                future = self.executor.submit(self.speak_and_play, text)
-                self.tts_queue.put(future)
             return False
         return True
 
@@ -293,6 +339,7 @@ class ConnectionHandler:
                 break
 
             end_time = time.time()
+
             self.logger.bind(tag=TAG).debug(
                 f"大模型返回时间: {end_time - start_time} 秒, 生成token={content}"
             )
@@ -316,6 +363,7 @@ class ConnectionHandler:
                     u_id=uuid_str, msg_type=MsgType.STOP_TTS_REQUEST, content=""
                 )
             )
+
         self.llm_finish_task = True
         self.dialogue.put(Message(role="assistant", content="".join(response_message)))
         self.logger.bind(tag=TAG).debug(
@@ -338,8 +386,9 @@ class ConnectionHandler:
             self.dialogue.put(Message(role="user", content=query))
 
         # Define intent functions
-        functions = self.func_handler.get_functions()
+        functions = None
-
+        if hasattr(self, "func_handler"):
+            functions = self.func_handler.get_functions()
         response_message = []
         processed_chars = 0  # 跟踪已处理的字符位置
 
@@ -378,6 +427,9 @@ class ConnectionHandler:
         msg_type = None
         for response in llm_responses:
             content, tools_call = response
+            if "content" in response:
+                content = response["content"]
+                tools_call = None
             if content is not None and len(content) > 0:
                 if len(response_message) <= 0 and (
                     content == "```" or "<tool_call>" in content
@@ -403,6 +455,7 @@ class ConnectionHandler:
                         break
 
                     end_time = time.time()
+
                     self.logger.bind(tag=TAG).debug(
                         f"大模型返回时间: {end_time - start_time} 秒, 生成token={content}"
                     )
@@ -465,9 +518,15 @@ class ConnectionHandler:
                     "id": function_id,
                     "arguments": function_arguments,
                 }
-                result = self.func_handler.handle_llm_function_call(
+
-                    self, function_call_data
+                # 处理MCP工具调用
-                )
+                if self.mcp_manager.is_mcp_tool(function_name):
+                    result = self._handle_mcp_tool_call(function_call_data)
+                else:
+                    # 处理系统函数
+                    result = self.func_handler.handle_llm_function_call(
+                        self, function_call_data
+                    )
                 self._handle_function_result(result, function_call_data, text_index + 1)
 
         # 存储对话内容
@@ -483,6 +542,48 @@ class ConnectionHandler:
 
         return True
 
+    def _handle_mcp_tool_call(self, function_call_data):
+        function_arguments = function_call_data["arguments"]
+        function_name = function_call_data["name"]
+        try:
+            args_dict = function_arguments
+            if isinstance(function_arguments, str):
+                try:
+                    args_dict = json.loads(function_arguments)
+                except json.JSONDecodeError:
+                    self.logger.bind(tag=TAG).error(
+                        f"无法解析 function_arguments: {function_arguments}"
+                    )
+                    return ActionResponse(
+                        action=Action.REQLLM, result="参数解析失败", response=""
+                    )
+
+            tool_result = asyncio.run_coroutine_threadsafe(
+                self.mcp_manager.execute_tool(function_name, args_dict), self.loop
+            ).result()
+            # meta=None content=[TextContent(type='text', text='北京当前天气:\n温度: 21°C\n天气: 晴\n湿度: 6%\n风向: 西北 风\n风力等级: 5级', annotations=None)] isError=False
+            content_text = ""
+            if tool_result is not None and tool_result.content is not None:
+                for content in tool_result.content:
+                    content_type = content.type
+                    if content_type == "text":
+                        content_text = content.text
+                    elif content_type == "image":
+                        pass
+
+            if len(content_text) > 0:
+                return ActionResponse(
+                    action=Action.REQLLM, result=content_text, response=""
+                )
+
+        except Exception as e:
+            self.logger.bind(tag=TAG).error(f"MCP工具调用错误: {e}")
+            return ActionResponse(
+                action=Action.REQLLM, result="工具调用出错", response=""
+            )
+
+        return ActionResponse(action=Action.REQLLM, result="工具调用出错", response="")
+
     def _handle_function_result(self, result, function_call_data, text_index):
         if result.action == Action.RESPONSE:  # 直接回复前端
             text = result.response
@@ -559,16 +660,44 @@ class ConnectionHandler:
             self.tts_first_text_index = text_index
         self.tts_last_text_index = text_index
 
-    async def close(self):
+    async def close(self, ws=None):
         """资源清理方法"""
-        # 清理其他资源
+        # 清理MCP资源
-        self.stop_event.set()
+        await self.mcp_manager.cleanup_all()
-        self.executor.shutdown(wait=False)
+
-        if self.websocket:
+        # 触发停止事件并清理资源
+        if self.stop_event:
+            self.stop_event.set()
+
+        # 立即关闭线程池
+        if self.executor:
+            self.executor.shutdown(wait=False, cancel_futures=True)
+            self.executor = None
+
+        # 清空任务队列
+        self._clear_queues()
+
+        if ws:
+            await ws.close()
+        elif self.websocket:
             await self.websocket.close()
         await self.tts.close()
         self.logger.bind(tag=TAG).info("连接资源已释放")
 
+    def _clear_queues(self):
+        # 清空所有任务队列
+        for q in [self.tts_queue, self.audio_play_queue]:
+            if not q:
+                continue
+            while not q.empty():
+                try:
+                    q.get_nowait()
+                except queue.Empty:
+                    continue
+            q.queue.clear()
+            # 添加毒丸信号到队列，确保线程退出
+            # q.queue.put(None)
+
     def reset_vad_states(self):
         self.client_audio_buffer = bytes()
         self.client_have_voice = False

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

@@ -1,22 +1,23 @@
-import asyncio
-from enum import Enum
-
 from config.logger import setup_logging
 import json
 from plugins_func.register import FunctionRegistry, ActionResponse, Action, ToolType
+from plugins_func.functions.hass_init import append_devices_to_prompt
+
 TAG = __name__
 logger = setup_logging()
 
 
 class FunctionHandler:
-    def __init__(self, config):
+    def __init__(self, conn):
-        self.config = config
+        self.conn = conn
+        self.config = conn.config
         self.function_registry = FunctionRegistry()
         self.register_nessary_functions()
         self.register_config_functions()
         self.functions_desc = self.function_registry.get_all_function_desc()
         func_names = self.current_support_functions()
         self.modify_plugin_loader_des(func_names)
+        self.finish_init = True
 
     def modify_plugin_loader_des(self, func_names):
         if "plugin_loader" not in func_names:
@@ -26,9 +27,11 @@ class FunctionHandler:
         func_names = ",".join(surport_plugins)
         for function_desc in self.functions_desc:
             if function_desc["function"]["name"] == "plugin_loader":
-                function_desc["function"]["description"] = function_desc["function"]["description"].replace("[plugins]", func_names)
+                function_desc["function"]["description"] = function_desc["function"][
+                    "description"
+                ].replace("[plugins]", func_names)
                 break
-        
+
     def upload_functions_desc(self):
         self.functions_desc = self.function_registry.get_all_function_desc()
 
@@ -47,16 +50,19 @@ class FunctionHandler:
     def register_nessary_functions(self):
         """注册必要的函数"""
         self.function_registry.register_function("handle_exit_intent")
-        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")
+        self.function_registry.register_function("get_lunar")
+        self.function_registry.register_function("handle_device")
 
     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"""
         for func in self.config["Intent"]["function_call"].get("functions", []):
             self.function_registry.register_function(func)
 
+        """home assistant需要初始化提示词"""
+        append_devices_to_prompt(self.conn)
+
     def get_function(self, name):
         return self.function_registry.get_function(name)
 
@@ -65,20 +71,27 @@ class FunctionHandler:
             function_name = function_call_data["name"]
             funcItem = self.get_function(function_name)
             if not funcItem:
-                return ActionResponse(action=Action.NOTFOUND, result="没有找到对应的函数", response="")
+                return ActionResponse(
+                    action=Action.NOTFOUND, result="没有找到对应的函数", response=""
+                )
             func = funcItem.func
             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 or funcItem.type == ToolType.IOT_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)
             elif funcItem.type == ToolType.CHANGE_SYS_PROMPT:
                 return func(conn, **arguments)
             else:
-                return ActionResponse(action=Action.NOTFOUND, result="没有找到对应的函数", response="")
+                return ActionResponse(
+                    action=Action.NOTFOUND, result="没有找到对应的函数", response=""
+                )
         except Exception as e:
             logger.bind(tag=TAG).error(f"处理function call错误: {e}")
 
-        return None
+        return None

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

@@ -1,8 +1,102 @@
 import json
 from config.logger import setup_logging
+from core.providers.tts.dto.dto import TTSMessageDTO, MsgType
+from core.handle.sendAudioHandle import send_stt_message
+from core.utils.util import remove_punctuation_and_length
+import shutil
+import asyncio
+import os
+import random
+import time
 
 logger = setup_logging()
 
+WAKEUP_CONFIG = {
+    "dir": "config/assets/",
+    "file_name": "wakeup_words",
+    "create_time": time.time(),
+    "refresh_time": 10,
+    "words": ["你好小智", "你好啊小智", "小智你好", "小智"],
+    "text": "",
+}
+
 
 async def handleHelloMessage(conn):
     await conn.websocket.send(json.dumps(conn.welcome_msg))
+
+
+async def checkWakeupWords(conn, text):
+    enable_wakeup_words_response_cache = conn.config[
+        "enable_wakeup_words_response_cache"
+    ]
+    """是否开启唤醒词加速"""
+    if not enable_wakeup_words_response_cache:
+        return False
+    """检查是否是唤醒词"""
+    _, text = remove_punctuation_and_length(text)
+    if text in conn.config.get("wakeup_words"):
+        await send_stt_message(conn, text)
+        conn.tts_first_text_index = 0
+        conn.tts_last_text_index = 0
+        conn.llm_finish_task = True
+
+        file = getWakeupWordFile(WAKEUP_CONFIG["file_name"])
+        if file is None:
+            asyncio.create_task(wakeupWordsResponse(conn))
+            return False
+        opus_packets, duration = conn.tts.audio_to_opus_data(file)
+        text_hello = WAKEUP_CONFIG["text"]
+        if not text_hello:
+            text_hello = text
+
+        conn.tts.tts_audio_queue.put(
+            TTSMessageDTO(
+                u_id=conn.u_id,
+                msg_type=MsgType.TTS_TEXT_RESPONSE,
+                content=opus_packets,
+                tts_finish_text="",
+                sentence_type=None,
+                duration=0,
+            )
+        )
+        if time.time() - WAKEUP_CONFIG["create_time"] > WAKEUP_CONFIG["refresh_time"]:
+            asyncio.create_task(wakeupWordsResponse(conn))
+        return True
+    return False
+
+
+def getWakeupWordFile(file_name):
+    for file in os.listdir(WAKEUP_CONFIG["dir"]):
+        if file.startswith("my_" + file_name):
+            """避免缓存文件是一个空文件"""
+            if os.stat(f"config/assets/{file}").st_size > (15 * 1024):
+                return f"config/assets/{file}"
+
+    """查找config/assets/目录下名称为wakeup_words的文件"""
+    for file in os.listdir(WAKEUP_CONFIG["dir"]):
+        if file.startswith(file_name):
+            return f"config/assets/{file}"
+    return None
+
+
+async def wakeupWordsResponse(conn):
+    """唤醒词响应"""
+    wakeup_word = random.choice(WAKEUP_CONFIG["words"])
+    result = conn.llm.response_no_stream(conn.config["prompt"], wakeup_word)
+    # TODO 将result转换为tts_message_dto
+    tts_file = None
+
+    if tts_file is not None and os.path.exists(tts_file):
+        file_type = os.path.splitext(tts_file)[1]
+        if file_type:
+            file_type = file_type.lstrip(".")
+        old_file = getWakeupWordFile("my_" + WAKEUP_CONFIG["file_name"])
+        if old_file is not None:
+            os.remove(old_file)
+        """将文件挪到"wakeup_words.mp3"""
+        shutil.move(
+            tts_file,
+            WAKEUP_CONFIG["dir"] + "my_" + WAKEUP_CONFIG["file_name"] + "." + file_type,
+        )
+        WAKEUP_CONFIG["create_time"] = time.time()
+        WAKEUP_CONFIG["text"] = result

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

@@ -3,6 +3,8 @@ import json
 import uuid
 from core.handle.sendAudioHandle import send_stt_message
 from core.utils.util import remove_punctuation_and_length
+from core.utils.dialogue import Message
+from loguru import logger
 
 TAG = __name__
 logger = setup_logging()
@@ -12,15 +14,20 @@ async def handle_user_intent(conn, text):
     # 检查是否有明确的退出命令
     if await check_direct_exit(conn, text):
         return True
+    # 4月4日因流式改造暂时关闭唤醒词加速功能
+    # # 检查是否是唤醒词
+    # if await checkWakeupWords(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)
+    intent_result = await analyze_intent_with_llm(conn, text)
-    if not intent:
+    if not intent_result:
         return False
     # 处理各种意图
-    return await process_intent_result(conn, intent, text)
+    return await process_intent_result(conn, intent_result, text)
 
 
 async def check_direct_exit(conn, text):
@@ -30,6 +37,7 @@ async def check_direct_exit(conn, text):
     for cmd in cmd_exit:
         if text == cmd:
             logger.bind(tag=TAG).info(f"识别到明确的退出命令: {text}")
+            await send_stt_message(conn, text)
             await conn.close()
             return True
     return False
@@ -37,7 +45,7 @@ async def check_direct_exit(conn, text):
 
 async def analyze_intent_with_llm(conn, text):
     """使用LLM分析用户意图"""
-    if not hasattr(conn, 'intent') or not conn.intent:
+    if not hasattr(conn, "intent") or not conn.intent:
         logger.bind(tag=TAG).warning("意图识别服务未初始化")
         return None
 
@@ -45,52 +53,64 @@ 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
+        return intent_result
-        try:
-            intent_data = json.loads(intent_result)
-            if "intent" in intent_data:
-                return intent_data["intent"]
-        except json.JSONDecodeError:
-            # 如果不是JSON格式，尝试直接获取意图文本
-            return intent_result.strip()
-
     except Exception as e:
         logger.bind(tag=TAG).error(f"意图识别失败: {str(e)}")
 
     return None
 
 
-async def process_intent_result(conn, intent, original_text):
+async def process_intent_result(conn, intent_result, original_text):
     """处理意图识别结果"""
-    # 处理退出意图
+    try:
-    if "结束聊天" in intent:
+        # 尝试将结果解析为JSON
-        logger.bind(tag=TAG).info(f"识别到退出意图: {intent}")
+        intent_data = json.loads(intent_result)
-        # 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)
-        return True
 
-    # 处理播放音乐意图
+        # 检查是否有function_call
-    if "播放音乐" in intent:
+        if "function_call" in intent_data:
-        logger.bind(tag=TAG).info(f"识别到音乐播放意图: {intent}")
+            # 直接从意图识别获取了function_call
-        # 调用play_music函数来播放音乐
+            logger.bind(tag=TAG).info(
-        song_name = extract_text_in_brackets(intent)
+                f"检测到function_call格式的意图结果: {intent_data['function_call']['name']}"
-        function_id = str(uuid.uuid4().hex)
+            )
-        function_name = "play_music"
+            function_name = intent_data["function_call"]["name"]
-        function_arguments = '{ "song_name": "' + song_name + '" }'
+            if function_name == "continue_chat":
+                return False
+            function_args = None
+            if "arguments" in intent_data["function_call"]:
+                function_args = intent_data["function_call"]["arguments"]
+            # 确保参数是字符串格式的JSON
+            if isinstance(function_args, dict):
+                function_args = json.dumps(function_args)
 
-        function_call_data = {
+            function_call_data = {
-            "name": function_name,
+                "name": function_name,
-            "id": function_id,
+                "id": str(uuid.uuid4().hex),
-            "arguments": function_arguments
+                "arguments": function_args,
-        }
+            }
-        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
+            await send_stt_message(conn, original_text)
 
-    # 默认返回False，表示继续常规聊天流程
+            # 使用executor执行函数调用和结果处理
-    return False
+            def process_function_call():
+                conn.dialogue.put(Message(role="user", content=original_text))
+                result = conn.func_handler.handle_llm_function_call(
+                    conn, function_call_data
+                )
+                if result and function_name != "play_music":
+                    # 获取当前最新的文本索引
+                    text = result.response
+                    if text is None:
+                        text = result.result
+                    if text is not None:
+                        # TODO 将text转换为tts_message_dto
+                        pass
+            # 将函数执行放在线程池中
+            conn.executor.submit(process_function_call)
+            return True
+        return False
+    except json.JSONDecodeError as e:
+        logger.bind(tag=TAG).error(f"处理意图结果时出错: {e}")
+        return False
 
 
 def extract_text_in_brackets(s):
@@ -100,10 +120,14 @@ def extract_text_in_brackets(s):
     :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('[')
+    left_bracket_index = s.find("[")
-    right_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:
+    if (
-        return s[left_bracket_index + 1:right_bracket_index]
+        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 ""
+        return ""

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

@@ -1,7 +1,13 @@
 import json
 import asyncio
 from config.logger import setup_logging
-from plugins_func.register import device_type_registry, register_function, ActionResponse, Action, ToolType
+from plugins_func.register import (
+    device_type_registry,
+    register_function,
+    ActionResponse,
+    Action,
+    ToolType,
+)
 
 TAG = __name__
 logger = setup_logging()
@@ -14,10 +20,13 @@ def wrap_async_function(async_func):
         try:
             # 获取连接对象（第一个参数）
             conn = args[0]
-            if not hasattr(conn, 'loop'):
+            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",
+                return ActionResponse(
-                                      "执行操作时出错: Connection对象没有loop属性")
+                    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
@@ -37,11 +46,14 @@ 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):
+    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}'")
+                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)
@@ -51,14 +63,15 @@ def create_iot_function(device_name, method_name, method_info):
             # 生成结果信息
             result = f"{device_name}的{method_name}操作执行成功"
 
-
             # 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))
+                    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:
@@ -86,7 +99,8 @@ def create_iot_query_function(device_name, prop_name, prop_info):
         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}'")
+                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)
 
@@ -126,7 +140,7 @@ class IotDescriptor:
         # 根据描述创建属性
         for key, value in properties.items():
             property_item = globals()[key] = {}
-            property_item['name'] = key
+            property_item["name"] = key
             property_item["description"] = value["description"]
             if value["type"] == "number":
                 property_item["value"] = 0
@@ -140,7 +154,7 @@ class IotDescriptor:
         for key, value in methods.items():
             method = globals()[key] = {}
             method["description"] = value["description"]
-            method['name'] = key
+            method["name"] = key
             for k, v in value["parameters"].items():
                 method[k] = {}
                 method[k]["description"] = v["description"]
@@ -177,19 +191,21 @@ def register_device_type(descriptor):
                     "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"
+                            "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}的{prop_info['description']}'"
+                            "description": f"查询失败时的友好回复，例如：'无法获取{device_name}的{prop_info['description']}'",
-                        }
+                        },
                     },
-                    "required": ["response_success", "response_failure"]
+                    "required": ["response_success", "response_failure"],
-                }
+                },
-            }
+            },
         }
         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)
+        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
@@ -200,22 +216,24 @@ def register_device_type(descriptor):
         parameters = {
             param_name: {
                 "type": param_info["type"],
-                "description": param_info["description"]
+                "description": param_info["description"],
             }
             for param_name, param_info in method_info["parameters"].items()
         }
 
         # 添加响应参数
-        parameters.update({
+        parameters.update(
-            "response_success": {
+            {
-                "type": "string",
+                "response_success": {
-                "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"
+                    "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",
+                "response_failure": {
-                "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"
+                    "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())
@@ -229,12 +247,14 @@ def register_device_type(descriptor):
                 "parameters": {
                     "type": "object",
                     "properties": parameters,
-                    "required": required_params
+                    "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)
+        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)
@@ -243,13 +263,24 @@ def register_device_type(descriptor):
 
 # 用于接受前端设备推送的搜索iot描述
 async def handleIotDescriptors(conn, descriptors):
+    wait_max_time = 5
+    while conn.func_handler is None or not conn.func_handler.finish_init:
+        await asyncio.sleep(1)
+        wait_max_time -= 1
+        if wait_max_time <= 0:
+            logger.bind(tag=TAG).error("连接对象没有func_handler")
+            return
     """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"],
+        iot_descriptor = IotDescriptor(
-                                       descriptor["methods"])
+            descriptor["name"],
+            descriptor["description"],
+            descriptor["properties"],
+            descriptor["methods"],
+        )
         conn.iot_descriptors[descriptor["name"]] = iot_descriptor
 
         if conn.use_function_call_mode:
@@ -258,18 +289,22 @@ async def handleIotDescriptors(conn, descriptors):
             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'):
+            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}")
+                    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'):
+    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}")
+        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):
@@ -281,11 +316,15 @@ async def handleIotStatus(conn, states):
                     for k, v in state["state"].items():
                         if property_item["name"] == k:
                             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")
+                                logger.bind(tag=TAG).error(
+                                    f"属性{property_item['name']}的值类型不匹配"
+                                )
                                 break
                             else:
                                 property_item["value"] = v
-                                logger.bind(tag=TAG).info(f"物联网状态更新: {key} , {property_item['name']} = {v}")
+                                logger.bind(tag=TAG).info(
+                                    f"物联网状态更新: {key} , {property_item['name']} = {v}"
+                                )
                             break
                 break
 
@@ -308,10 +347,14 @@ async def set_iot_status(conn, name, property_name, value):
             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")
+                        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}")
+                    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}")
 
@@ -324,15 +367,19 @@ async def send_iot_conn(conn, name, method_name, parameters):
             for method in value.methods:
                 # 找到了方法
                 if method["name"] == method_name:
-                    await conn.websocket.send(json.dumps({
+                    await conn.websocket.send(
-                        "type": "iot",
+                        json.dumps(
-                        "commands": [
                             {
-                                "name": name,
+                                "type": "iot",
-                                "method": method_name,
+                                "commands": [
-                                "parameters": parameters
+                                    {
+                                        "name": name,
+                                        "method": method_name,
+                                        "parameters": parameters,
+                                    }
+                                ],
                             }
-                        ]
+                        )
-                    }))
+                    )
                     return
     logger.bind(tag=TAG).error(f"未找到方法{method_name}")

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

@@ -21,7 +21,9 @@ async def handleAudioMessage(conn, audio):
     if have_voice == False and conn.client_have_voice == False:
         await no_voice_close_connect(conn)
         conn.asr_audio.append(audio)
-        conn.asr_audio = conn.asr_audio[-5:]  # 保留最新的5帧音频内容，解决ASR句首丢字问题
+        conn.asr_audio = conn.asr_audio[
+            -10:
+        ]  # 保留最新的10帧音频内容，解决ASR句首丢字问题
         return
     conn.client_no_voice_last_time = 0.0
     conn.asr_audio.append(audio)
@@ -30,10 +32,12 @@ async def handleAudioMessage(conn, audio):
         conn.client_abort = False
         conn.asr_server_receive = False
         # 音频太短了，无法识别
-        if len(conn.asr_audio) < 10:
+        if len(conn.asr_audio) < 15:
             conn.asr_server_receive = True
         else:
-            text, file_path = await conn.asr.speech_to_text(conn.asr_audio, conn.session_id)
+            text, file_path = await conn.asr.speech_to_text(
+                conn.asr_audio, conn.session_id
+            )
             logger.bind(tag=TAG).info(f"识别文本: {text}")
             text_len, _ = remove_punctuation_and_length(text)
             if text_len > 0:
@@ -47,12 +51,12 @@ async def handleAudioMessage(conn, audio):
 async def startToChat(conn, text):
     # 首先进行意图分析
     intent_handled = await handle_user_intent(conn, text)
-    
+
     if intent_handled:
         # 如果意图已被处理，不再进行聊天
         conn.asr_server_receive = True
         return
-    
+
     # 意图未被处理，继续常规聊天流程
     await send_stt_message(conn, text)
     if conn.use_function_call_mode:
@@ -67,10 +71,17 @@ async def no_voice_close_connect(conn):
         conn.client_no_voice_last_time = time.time() * 1000
     else:
         no_voice_time = time.time() * 1000 - conn.client_no_voice_last_time
-        close_connection_no_voice_time = conn.config.get("close_connection_no_voice_time", 120)
+        close_connection_no_voice_time = conn.config.get(
-        if not conn.close_after_chat and no_voice_time > 1000 * close_connection_no_voice_time:
+            "close_connection_no_voice_time", 120
+        )
+        if (
+            not conn.close_after_chat
+            and no_voice_time > 1000 * close_connection_no_voice_time
+        ):
             conn.close_after_chat = True
             conn.client_abort = False
             conn.asr_server_receive = False
-            prompt = "请你以“时间过得真快”未来头，用富有感情、依依不舍的话来结束这场对话吧。"
+            prompt = (
+                "请你以“时间过得真快”未来头，用富有感情、依依不舍的话来结束这场对话吧。"
+            )
             await startToChat(conn, prompt)

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

@@ -2,8 +2,11 @@ from config.logger import setup_logging
 import json
 from core.handle.abortHandle import handleAbortMessage
 from core.handle.helloHandle import handleHelloMessage
+from core.utils.util import remove_punctuation_and_length
 from core.handle.receiveAudioHandle import startToChat, handleAudioMessage
+from core.handle.sendAudioHandle import send_stt_message, send_tts_message
 from core.handle.iotHandle import handleIotDescriptors, handleIotStatus
+import asyncio
 
 TAG = __name__
 logger = setup_logging()
@@ -32,17 +35,31 @@ async def handleTextMessage(conn, message):
                 conn.client_have_voice = True
                 conn.client_voice_stop = True
                 if len(conn.asr_audio) > 0:
-                    await handleAudioMessage(conn, b'')
+                    await handleAudioMessage(conn, b"")
             elif msg_json["state"] == "detect":
                 conn.asr_server_receive = False
                 conn.client_have_voice = False
                 conn.asr_audio.clear()
                 if "text" in msg_json:
-                    await startToChat(conn, msg_json["text"])
+                    text = msg_json["text"]
+                    _, text = remove_punctuation_and_length(text)
+
+                    # 识别是否是唤醒词
+                    is_wakeup_words = text in conn.config.get("wakeup_words")
+                    # 是否开启唤醒词回复
+                    enable_greeting = conn.config.get("enable_greeting", True)
+
+                    if is_wakeup_words and not enable_greeting:
+                        # 如果是唤醒词，且关闭了唤醒词回复，就不用回答
+                        await send_stt_message(conn, text)
+                        await send_tts_message(conn, "stop", None)
+                    else:
+                        # 否则需要LLM对文字内容进行答复
+                        await startToChat(conn, text)
         elif msg_json["type"] == "iot":
             if "descriptors" in msg_json:
-                await handleIotDescriptors(conn, msg_json["descriptors"])
+                asyncio.create_task(handleIotDescriptors(conn, msg_json["descriptors"]))
             if "states" in msg_json:
-                await handleIotStatus(conn, msg_json["states"])
+                asyncio.create_task(handleIotStatus(conn, msg_json["states"]))
     except json.JSONDecodeError:
         await conn.websocket.send(message)

main/xiaozhi-server/core/mcp/MCPClient.py

@@ -0,0 +1,86 @@
+from datetime import timedelta
+from typing import Optional
+from contextlib import AsyncExitStack
+import os, shutil
+from mcp import ClientSession, StdioServerParameters
+from mcp.client.stdio import stdio_client
+
+from config.logger import setup_logging
+
+TAG = __name__
+
+class MCPClient:
+    def __init__(self, config):
+        # Initialize session and client objects
+        self.session: Optional[ClientSession] = None
+        self.exit_stack = AsyncExitStack()
+        self.logger = setup_logging()
+        self.config = config
+        self.tolls = []
+
+    async def initialize(self):
+        args = self.config.get("args", [])
+
+        command = (
+            shutil.which("npx")
+            if self.config["command"] == "npx"
+            else self.config["command"]
+        )
+        
+        env={**os.environ}
+        if self.config.get("env"):
+            env.update(self.config["env"])
+        
+        server_params = StdioServerParameters(
+            command=command,
+            args=args,
+            env=env
+        )
+        
+        stdio_transport = await self.exit_stack.enter_async_context(stdio_client(server_params))
+        self.stdio, self.write = stdio_transport
+        time_out_delta =  timedelta(seconds=15)
+        self.session = await self.exit_stack.enter_async_context(ClientSession(read_stream=self.stdio, write_stream=self.write, read_timeout_seconds=time_out_delta))
+        
+        await self.session.initialize()
+        
+        # List available tools
+        response = await self.session.list_tools()
+        tools = response.tools
+        self.tools = tools
+        self.logger.bind(tag=TAG).info(f"Connected to server with tools:{[tool.name for tool in tools]}")
+    
+    def has_tool(self, tool_name):
+        return any(tool.name == tool_name for tool in self.tools)
+    
+    def get_available_tools(self):
+        available_tools = [{"type": "function", "function":{ 
+            "name": tool.name,
+            "description": tool.description,
+            "parameters": tool.inputSchema
+        } } for tool in self.tools]
+
+        return available_tools
+    
+    async def call_tool(self, tool_name: str, tool_args: dict):
+        self.logger.bind(tag=TAG).info(f"MCPClient Calling tool {tool_name} with args: {tool_args}")
+        try:
+            response = await self.session.call_tool(tool_name, tool_args)
+        except Exception as e:
+            self.logger.bind(tag=TAG).error(f"Error calling tool {tool_name}: {e}")
+            from types import SimpleNamespace
+            error_content = SimpleNamespace(
+                type='text',
+                text=f"Error calling tool {tool_name}: {e}"
+            )
+            error_response = SimpleNamespace(
+                content=[error_content],
+                isError=True
+            )
+            return error_response
+        self.logger.bind(tag=TAG).info(f"MCPClient Response from tool {tool_name}: {response}")
+        return response
+
+    async def cleanup(self):
+        """Clean up resources"""
+        await self.exit_stack.aclose()

main/xiaozhi-server/core/mcp/manager.py

@@ -0,0 +1,110 @@
+"""MCP服务管理器"""
+import os, json
+from typing import Dict, Any, List
+from .MCPClient import MCPClient
+from config.logger import setup_logging
+from core.utils.util import get_project_dir
+from plugins_func.register import register_function, ActionResponse, Action, ToolType
+
+TAG = __name__
+
+class MCPManager:
+    """管理多个MCP服务的集中管理器"""
+
+    def __init__(self,conn) -> None:
+        """
+        初始化MCP管理器
+        """
+        self.conn = conn
+        self.logger = setup_logging()
+        self.config_path = get_project_dir() + 'data/.mcp_server_settings.json'
+        if os.path.exists(self.config_path) == False:
+            self.config_path = ""
+            self.logger.bind(tag=TAG).warning(f"请检查mcp服务配置文件：data/.mcp_server_settings.json")
+        self.client: Dict[str, MCPClient] = {}
+        self.tools = []
+
+    def load_config(self) -> Dict[str, Any]:
+        """加载MCP服务配置
+        Returns:
+            Dict[str, Any]: 服务配置字典
+        """
+        if len(self.config_path) == 0:
+            return {}
+        
+        try:
+            with open(self.config_path, 'r', encoding='utf-8') as f:
+                config = json.load(f)
+            return config.get('mcpServers', {})
+        except Exception as e:
+            self.logger.bind(tag=TAG).error(f"Error loading MCP config from {self.config_path}: {e}")
+            return {}
+
+    async def initialize_servers(self) -> None:
+        """初始化所有MCP服务"""
+        config = self.load_config()
+        for name, srv_config in config.items():
+            if not srv_config.get("command"):
+                self.logger.bind(tag=TAG).warning(f"Skipping server {name}: command not specified")
+                continue
+
+            try:
+                client = MCPClient(srv_config)
+                await client.initialize()
+                self.client[name] = client
+                self.logger.bind(tag=TAG).info(f"Initialized MCP client: {name}")
+                client_tools = client.get_available_tools()
+                self.tools.extend(client_tools)
+                for tool in client_tools:
+                    func_name = "mcp_"+tool["function"]["name"]
+                    register_function(func_name, tool, ToolType.MCP_CLIENT)(self.execute_tool)
+                    self.conn.func_handler.function_registry.register_function(func_name)
+
+            except Exception as e:
+                self.logger.bind(tag=TAG).error(f"Failed to initialize MCP server {name}: {e}")
+        self.conn.func_handler.upload_functions_desc()
+
+    def get_all_tools(self) -> List[Dict[str, Any]]:
+        """获取所有服务的工具function定义
+        Returns:
+            List[Dict[str, Any]]: 所有工具的function定义列表
+        """
+        return self.tools
+
+    def is_mcp_tool(self, tool_name: str) -> bool:
+        """检查是否是MCP工具
+        Args:
+            tool_name: 工具名称
+        Returns:
+            bool: 是否是MCP工具
+        """
+        for tool in self.tools:
+            if tool.get("function") != None and tool["function"].get("name") == tool_name:
+                return True
+        return False
+
+    async def execute_tool(self, tool_name: str, arguments: Dict[str, Any]) -> Any:
+        """执行工具调用
+        Args:
+            tool_name: 工具名称
+            arguments: 工具参数 
+        Returns:
+            Any: 工具执行结果
+        Raises:
+            ValueError: 工具未找到时抛出
+        """
+        self.logger.bind(tag=TAG).info(f"Executing tool {tool_name} with arguments: {arguments}")
+        for client in self.client.values():
+            if client.has_tool(tool_name):
+                return await client.call_tool(tool_name, arguments)
+            
+        raise ValueError(f"Tool {tool_name} not found in any MCP server")
+
+    async def cleanup_all(self) -> None:
+        for name, client in self.client.items():
+            try:
+                await client.cleanup()
+                self.logger.bind(tag=TAG).info(f"Cleaned up MCP client: {name}")
+            except Exception as e:
+                self.logger.bind(tag=TAG).error(f"Error cleaning up MCP client {name}: {e}")
+        self.client.clear()

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

@@ -10,14 +10,21 @@ class IntentProviderBase(ABC):
     def __init__(self, config):
         self.config = config
         self.intent_options = config.get("intent_options", {
+            "handle_exit_intent": "结束聊天, 用户发来如再见之类的表示结束的话, 不想再进行对话的时候",
+            "play_music": "播放音乐, 用户希望你可以播放音乐, 只用于播放音乐的意图",
+            "get_weather": "查询天气, 用户希望查询某个地点的天气情况",
+            "get_news": "查询新闻, 用户希望查询最新新闻或特定类型的新闻",
+            "get_lunar": "用于获取今天的阴历/农历和黄历信息",
+            "get_time": "获取今天日期或者当前时间信息",
             "continue_chat": "继续聊天",
-            "end_chat": "结束聊天",
-            "play_music": "播放音乐"
         })
 
     def set_llm(self, llm):
         self.llm = llm
-        logger.bind(tag=TAG).debug("Set LLM for intent provider")
+        # 获取模型名称和类型信息
+        model_name = getattr(llm, 'model_name', str(llm.__class__.__name__))
+        # 记录更详细的日志
+        logger.bind(tag=TAG).info(f"意图识别设置LLM: {model_name}")
 
     @abstractmethod
     async def detect_intent(self, conn, dialogue_history: List[Dict], text: str) -> str:
@@ -30,5 +37,6 @@ class IntentProviderBase(ABC):
             - "继续聊天"
             - "结束聊天" 
             - "播放音乐 歌名" 或 "随机播放音乐"
+            - "查询天气 地点名" 或 "查询天气 [当前位置]"
         """
         pass

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

@@ -0,0 +1,20 @@
+from ..base import IntentProviderBase
+from typing import List, Dict
+from config.logger import setup_logging
+
+TAG = __name__
+logger = setup_logging()
+
+
+class IntentProvider(IntentProviderBase):
+    async def detect_intent(self, conn, dialogue_history: List[Dict], text: str) -> str:
+        """
+        默认的意图识别实现，始终返回继续聊天
+        Args:
+            dialogue_history: 对话历史记录列表
+            text: 本次对话记录
+        Returns:
+            固定返回"继续聊天"
+        """
+        logger.bind(tag=TAG).debug("Using functionCallProvider, always returning continue chat")
+        return self.intent_options["continue_chat"]

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

@@ -3,6 +3,9 @@ from ..base import IntentProviderBase
 from plugins_func.functions.play_music import initialize_music_handler
 from config.logger import setup_logging
 import re
+import json
+import hashlib
+import time
 
 TAG = __name__
 logger = setup_logging()
@@ -13,6 +16,10 @@ class IntentProvider(IntentProviderBase):
         super().__init__(config)
         self.llm = None
         self.promot = self.get_intent_system_prompt()
+        # 添加缓存管理
+        self.intent_cache = {}  # 缓存意图识别结果
+        self.cache_expiry = 600  # 缓存有效期10分钟
+        self.cache_max_size = 100  # 最多缓存100个意图
 
     def get_intent_system_prompt(self) -> str:
         """
@@ -22,62 +29,119 @@ class IntentProvider(IntentProviderBase):
         """
         intent_list = []
 
-        """
-        "continue_chat":    "1.继续聊天, 除了播放音乐和结束聊天的时候的选项, 比如日常的聊天和问候, 对话等",
-        "end_chat":         "2.结束聊天, 用户发来如再见之类的表示结束的话, 不想再进行对话的时候",
-        "play_music":       "3.播放音乐, 用户希望你可以播放音乐, 只用于播放音乐的意图"
-        """
-        for key, value in self.intent_options.items():
-            if key == "play_music":
-                intent_list.append("3.播放音乐, 用户希望你可以播放音乐, 只用于播放音乐的意图")
-            elif key == "end_chat":
-                intent_list.append("2.结束聊天, 用户发来如再见之类的表示结束的话, 不想再进行对话的时候")
-            elif key == "continue_chat":
-                intent_list.append("1.继续聊天, 除了播放音乐和结束聊天的时候的选项, 比如日常的聊天和问候, 对话等")
-            else:
-                intent_list.append(value)
-
-        # "如果是唱歌、听歌、播放音乐，请指定歌名，格式为'播放音乐 [识别出的歌名]'。\n"
-        # "如果听不出具体歌名，可以返回'随机播放音乐'。\n"
-        # "只需要返回意图结果的json，不要解释。"
-        # "返回格式如下：\n"
         prompt = (
-            "你是一个意图识别助手。你需要根据和用户的对话记录，重点分析用户的最后一句话，判断用户意图属于以下哪一类(使用<start>和<end>标志)：\n"
+            "你是一个意图识别助手。请分析用户的最后一句话，判断用户意图属于以下哪一类：\n"
             "<start>"
             f"{', '.join(intent_list)}"
             "<end>\n"
-            "你需要按照以下的步骤处理用户的对话"
+            "处理步骤:"
-            "1. 思考出对话的意图是哪一类的"
+            "1. 思考意图类型，生成function_call格式"
-            "2. 属于1和2的意图, 直接返回，返回格式如下：\n"
+            "\n\n"
-            "{intent: '用户意图'}\n"
+            "返回格式示例：\n"
-            "3. 属于3的意图，则继续分析用户希望播放的音乐\n"
+            "1. 播放音乐意图: {\"function_call\": {\"name\": \"play_music\", \"arguments\": {\"song_name\": \"音乐名称\"}}}\n"
-            "4. 如果无法识别出具体歌名，可以返回'随机播放音乐'\n"
+            "2. 查询天气意图: {\"function_call\": {\"name\": \"get_weather\", \"arguments\": {\"location\": \"地点名称\", \"lang\": \"zh_CN\"}}}\n"
-            "{intent: '播放音乐 [获取的音乐名字]'}\n"
+            "3. 查询新闻意图: {\"function_call\": {\"name\": \"get_news\", \"arguments\": {\"category\": \"新闻类别\", \"detail\": false, \"lang\": \"zh_CN\"}}}\n"
-            "下面是几个处理的示例(思考的内容不返回, 只返回json部分, 无额外的内容)\n"
+            "4. 结束对话意图: {\"function_call\": {\"name\": \"handle_exit_intent\", \"arguments\": {\"say_goodbye\": \"goodbye\"}}}\n"
-            "```"
+            "5. 获取当天日期时间: {\"function_call\": {\"name\": \"get_time\"}}\n"
+            "6. 获取当前黄历意图: {\"function_call\": {\"name\": \"get_lunar\"}}\n"
+            "7. 继续聊天意图: {\"function_call\": {\"name\": \"continue_chat\"}}\n"
+            "\n"
+            "注意:\n"
+            "- 播放音乐：无歌名时，song_name设为\"random\"\n"
+            "- 查询天气：无地点时，location设为null\n"
+            "- 查询新闻：无类别时，category设为null；查询详情时，detail设为true\n"
+            "- 如果没有明显的意图，应按照继续聊天意图处理\n"
+            "- 只返回纯JSON，不要任何其他内容\n"
+            "\n"
+            "示例分析:\n"
+            "```\n"
+            "用户: 你好小智\n"
+            "返回: {\"function_call\": {\"name\": \"continue_chat\"}}\n"
+            "```\n"
+            "```\n"
             "用户: 你今天怎么样?\n"
-            "思考(不返回): 用户发来的数据是一个问候语，属于继续聊天的意图, 是种类1, 种类1的需求是直接返回\n"
+            "返回: {\"function_call\": {\"name\": \"continue_chat\"}}\n"
-            "返回结果: {intent: '继续聊天'}\n"
+            "```\n"
-            "```"
+            "```\n"
-            "用户: 我今天有点累了, 我们明天再聊吧\n"
+            "用户: 现在是几号了?现在几点了？\n"
-            "思考(不返回): 用户表达了今天不想继续对话，属于结束聊天的意图, 是种类2, 种类2的需求是直接返回\n"
+            "返回: {\"function_call\": {\"name\": \"get_time\"}}\n"
-            "返回结果: {intent: '结束聊天'}\n"
+            "```\n"
-            "```"
+            "```\n"
-            "用户: 我今天有点累了, 我们明天再聊吧\n"
+            "用户: 今天农历是多少？\n"
-            "思考(不返回): 用户表达了今天不想继续对话，属于结束聊天的意图, 是种类2, 种类2的需求是直接返回\n"
+            "返回: {\"function_call\": {\"name\": \"get_lunar\"}}\n"
-            "返回结果: {intent: '结束聊天'}\n"
+            "```\n"
-            "```"
+            "```\n"
-            "用户: 你可以播放一首中秋月给我听吗\n"
+            "用户: 我们明天再聊吧\n"
-            "思考(不返回): 用户表达了想听音乐的续签，属于播放音乐的意图, 是种类3, 种类3的需求需要继续判断播放的音乐, 这里用户希望的歌曲名明确给出是中秋月\n"
+            "返回: {\"function_call\": {\"name\": \"handle_exit_intent\"}}\n"
-            "返回结果: {intent: '播放音乐 [中秋月]'}\n"
+            "```\n"
-            "```"
+            "```\n"
-            "你现在可以使用的音乐的名称如下(使用<start>和<end>标志):\n"
+            "用户: 播放中秋月\n"
+            "返回: {\"function_call\": {\"name\": \"play_music\", \"arguments\": {\"song_name\": \"中秋月\"}}}\n"
+            "```\n"
+            "```\n"
+            "用户: 北京天气怎么样\n"
+            "返回: {\"function_call\": {\"name\": \"get_weather\", \"arguments\": {\"location\": \"北京\", \"lang\": \"zh_CN\"}}}\n"
+            "```\n"
+            "```\n"
+            "用户: 今天天气怎么样\n"
+            "返回: {\"function_call\": {\"name\": \"get_weather\", \"arguments\": {\"location\": null, \"lang\": \"zh_CN\"}}}\n"
+            "```\n"
+            "```\n"
+            "用户: 播报财经新闻\n"
+            "返回: {\"function_call\": {\"name\": \"get_news\", \"arguments\": {\"category\": \"财经\", \"detail\": false, \"lang\": \"zh_CN\"}}}\n"
+            "```\n"
+            "```\n"
+            "用户: 有什么最新新闻\n"
+            "返回: {\"function_call\": {\"name\": \"get_news\", \"arguments\": {\"category\": null, \"detail\": false, \"lang\": \"zh_CN\"}}}\n"
+            "```\n"
+            "```\n"
+            "用户: 详细介绍一下这条新闻\n"
+            "返回: {\"function_call\": {\"name\": \"get_news\", \"arguments\": {\"detail\": true, \"lang\": \"zh_CN\"}}}\n"
+            "```\n"
+            "可用的音乐名称:\n"
         )
         return prompt
+        
+    def clean_cache(self):
+        """清理过期缓存"""
+        now = time.time()
+        # 找出过期键
+        expired_keys = [k for k, v in self.intent_cache.items() if now - v['timestamp'] > self.cache_expiry]
+        for key in expired_keys:
+            del self.intent_cache[key]
+            
+        # 如果缓存太大，移除最旧的条目
+        if len(self.intent_cache) > self.cache_max_size:
+            # 按时间戳排序并保留最新的条目
+            sorted_items = sorted(self.intent_cache.items(), key=lambda x: x[1]['timestamp'])
+            for key, _ in sorted_items[:len(sorted_items) - self.cache_max_size]:
+                del self.intent_cache[key]
 
     async def detect_intent(self, conn, dialogue_history: List[Dict], text: str) -> str:
         if not self.llm:
             raise ValueError("LLM provider not set")
+            
+        # 记录整体开始时间
+        total_start_time = time.time()
+            
+        # 打印使用的模型信息
+        model_info = getattr(self.llm, 'model_name', str(self.llm.__class__.__name__))
+        logger.bind(tag=TAG).info(f"使用意图识别模型: {model_info}")
+            
+        # 计算缓存键
+        cache_key = hashlib.md5(text.encode()).hexdigest()
+        
+        # 检查缓存
+        if cache_key in self.intent_cache:
+            cache_entry = self.intent_cache[cache_key]
+            # 检查缓存是否过期
+            if time.time() - cache_entry['timestamp'] <= self.cache_expiry:
+                cache_time = time.time() - total_start_time
+                logger.bind(tag=TAG).info(f"使用缓存的意图: {cache_key} -> {cache_entry['intent']}, 耗时: {cache_time:.4f}秒")
+                return cache_entry['intent']
+                
+        # 清理缓存
+        self.clean_cache()
 
         # 构建用户最后一句话的提示
         msgStr = ""
@@ -94,18 +158,78 @@ class IntentProvider(IntentProviderBase):
         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}")
+        
+        # 记录预处理完成时间
+        preprocess_time = time.time() - total_start_time
+        logger.bind(tag=TAG).debug(f"意图识别预处理耗时: {preprocess_time:.4f}秒")
+        
         # 使用LLM进行意图识别
+        llm_start_time = time.time()
+        logger.bind(tag=TAG).info(f"开始LLM意图识别调用, 模型: {model_info}")
+        
         intent = self.llm.response_no_stream(
             system_prompt=prompt_music,
             user_prompt=user_prompt
         )
-        # 使用正则表达式提取大括号中的内容  
+        
-        # 使用正则表达式提取 {} 中的内容
+        # 记录LLM调用完成时间
-        match = re.search(r'\{.*?\}', intent)
+        llm_time = time.time() - llm_start_time
+        logger.bind(tag=TAG).info(f"LLM意图识别完成, 模型: {model_info}, 调用耗时: {llm_time:.4f}秒")
+        
+        # 记录后处理开始时间
+        postprocess_start_time = time.time()
+        
+        # 清理和解析响应
+        intent = intent.strip()
+        # 尝试提取JSON部分
+        match = re.search(r'\{.*\}', intent, re.DOTALL)
         if match:
-            result = match.group(0)
+            intent = match.group(0)
-            intent = result
+        
-        else:
+        # 记录总处理时间
-            intent = "{intent: '继续聊天'}"
+        total_time = time.time() - total_start_time
-        logger.bind(tag=TAG).info(f"Detected intent: {intent}")
+        logger.bind(tag=TAG).info(f"【意图识别性能】模型: {model_info}, 总耗时: {total_time:.4f}秒, LLM调用: {llm_time:.4f}秒, 查询: '{text[:20]}...'")
-        return intent.strip()
+        
+        # 尝试解析为JSON
+        try:
+            intent_data = json.loads(intent)
+            # 如果包含function_call，则格式化为适合处理的格式
+            if "function_call" in intent_data:
+                function_data = intent_data["function_call"]
+                function_name = function_data.get("name")
+                function_args = function_data.get("arguments", {})
+                
+                # 记录识别到的function call
+                logger.bind(tag=TAG).info(f"识别到function call: {function_name}, 参数: {function_args}")
+                
+                # 添加到缓存
+                self.intent_cache[cache_key] = {
+                    'intent': intent,
+                    'timestamp': time.time()
+                }
+                
+                # 后处理时间
+                postprocess_time = time.time() - postprocess_start_time
+                logger.bind(tag=TAG).debug(f"意图后处理耗时: {postprocess_time:.4f}秒")
+                
+                # 确保返回完全序列化的JSON字符串
+                return intent
+            else:
+                # 添加到缓存
+                self.intent_cache[cache_key] = {
+                    'intent': intent,
+                    'timestamp': time.time()
+                }
+                
+                # 后处理时间
+                postprocess_time = time.time() - postprocess_start_time
+                logger.bind(tag=TAG).debug(f"意图后处理耗时: {postprocess_time:.4f}秒")
+                
+                # 返回普通意图
+                return intent
+        except json.JSONDecodeError:
+            # 后处理时间
+            postprocess_time = time.time() - postprocess_start_time
+            logger.bind(tag=TAG).error(f"无法解析意图JSON: {intent}, 后处理耗时: {postprocess_time:.4f}秒")
+            # 如果解析失败，默认返回继续聊天意图
+            return "{\"intent\": \"继续聊天\"}"

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

@@ -0,0 +1,52 @@
+from config.logger import setup_logging
+from http import HTTPStatus
+from dashscope import Application
+from core.providers.llm.base import LLMProviderBase
+
+TAG = __name__
+logger = setup_logging()
+
+class LLMProvider(LLMProviderBase):
+    def __init__(self, config):
+        self.api_key = config["api_key"]
+        self.app_id = config["app_id"]
+        self.base_url = config.get("base_url")
+        self.is_No_prompt = config.get("is_no_prompt")
+        self.memory_id = config.get("ali_memory_id")
+
+    def response(self, session_id, dialogue):
+        try:
+            # 处理dialogue
+            if self.is_No_prompt:
+                dialogue.pop(0)
+                logger.bind(tag=TAG).debug(f"【阿里百练API服务】处理后的dialogue: {dialogue}")
+
+            # 构造调用参数
+            call_params = {
+                "api_key": self.api_key,
+                "app_id": self.app_id,
+                "session_id": session_id,
+                "messages": dialogue
+            }
+            if self.memory_id != False:
+                # 百练memory需要prompt参数
+                prompt = dialogue[-1].get("content")
+                call_params["memory_id"] = self.memory_id
+                call_params["prompt"] = prompt
+                logger.bind(tag=TAG).debug(f"【阿里百练API服务】处理后的prompt: {prompt}")
+
+            responses = Application.call(**call_params)
+            if responses.status_code != HTTPStatus.OK:
+                logger.bind(tag=TAG).error(
+                    f"code={responses.status_code}, "
+                    f"message={responses.message}, "
+                    f"请参考文档：https://help.aliyun.com/zh/model-studio/developer-reference/error-code"
+                )
+                yield "【阿里百练API服务响应异常】"
+            else:
+                logger.bind(tag=TAG).debug(f"【阿里百练API服务】构造参数: {call_params}")
+                yield responses.output.text
+
+        except Exception as e:
+            logger.bind(tag=TAG).error(f"【阿里百练API服务】响应异常: {e}")
+            yield "【LLM服务响应异常】"

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

@@ -6,11 +6,12 @@ from core.providers.llm.base import LLMProviderBase
 TAG = __name__
 logger = setup_logging()
 
+
 class LLMProvider(LLMProviderBase):
     def __init__(self, config):
         self.api_key = config["api_key"]
         self.mode = config.get("mode", "chat-messages")
-        self.base_url = config.get("base_url", "https://api.dify.ai/v1").rstrip('/')
+        self.base_url = config.get("base_url", "https://api.dify.ai/v1").rstrip("/")
         self.session_conversation_map = {}  # 存储session_id和conversation_id的映射
 
     def response(self, session_id, dialogue):
@@ -22,57 +23,58 @@ class LLMProvider(LLMProviderBase):
             # 发起流式请求
             if self.mode == "chat-messages":
                 request_json = {
-                        "query": last_msg["content"],
+                    "query": last_msg["content"],
-                        "response_mode": "streaming",
+                    "response_mode": "streaming",
-                        "user": session_id,
+                    "user": session_id,
-                        "inputs": {},
+                    "inputs": {},
-                        "conversation_id": conversation_id
+                    "conversation_id": conversation_id,
-                    }
+                }
             elif self.mode == "workflows/run":
                 request_json = {
                     "inputs": {"query": last_msg["content"]},
                     "response_mode": "streaming",
-                    "user": session_id
+                    "user": session_id,
                 }
             elif self.mode == "completion-messages":
                 request_json = {
                     "inputs": {"query": last_msg["content"]},
                     "response_mode": "streaming",
-                    "user": session_id
+                    "user": session_id,
                 }
 
             with requests.post(
-                    f"{self.base_url}/{self.mode}",
+                f"{self.base_url}/{self.mode}",
-                    headers={"Authorization": f"Bearer {self.api_key}"},
+                headers={"Authorization": f"Bearer {self.api_key}"},
-                    json=request_json,
+                json=request_json,
-                    stream=True
+                stream=True,
             ) as r:
                 if self.mode == "chat-messages":
                     for line in r.iter_lines():
-                        if line.startswith(b'data: '):
+                        if line.startswith(b"data: "):
                             event = json.loads(line[6:])
                             # 如果没有找到conversation_id，则获取此次conversation_id
                             if not conversation_id:
-                                conversation_id = event.get('conversation_id')
+                                conversation_id = event.get("conversation_id")
-                                self.session_conversation_map[session_id] = conversation_id  # 更新映射
+                                self.session_conversation_map[session_id] = (
-                            if event.get('answer'):
+                                    conversation_id  # 更新映射
-                                yield event['answer']
+                                )
+                            if event.get("answer"):
+                                yield event["answer"]
                 elif self.mode == "workflows/run":
                     for line in r.iter_lines():
-                        # logger.bind(tag=TAG).info(f"chat message response: {line}")
+                        if line.startswith(b"data: "):
-                        if line.startswith(b'data: '):
                             event = json.loads(line[6:])
-                            if event.get('event') == "workflow_finished":
+                            if event.get("event") == "workflow_finished":
-                                if event['data']['status'] == "succeeded":
+                                if event["data"]["status"] == "succeeded":
-                                    yield event['data']['outputs']['answer']
+                                    yield event["data"]["outputs"]["answer"]
                                 else:
                                     yield "【服务响应异常】"
                 elif self.mode == "completion-messages":
                     for line in r.iter_lines():
-                        if line.startswith(b'data: '):
+                        if line.startswith(b"data: "):
                             event = json.loads(line[6:])
-                            if event.get('answer'):
+                            if event.get("answer"):
-                                yield event['answer'] 
+                                yield event["answer"]
 
         except Exception as e:
             logger.bind(tag=TAG).error(f"Error in response generation: {e}")

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

@@ -1,62 +0,0 @@
-import requests
-from requests.exceptions import RequestException
-from config.logger import setup_logging
-from core.providers.llm.base import LLMProviderBase
-
-TAG = __name__
-logger = setup_logging()
-
-
-class LLMProvider(LLMProviderBase):
-    def __init__(self, config):
-        self.agent_id = config.get("agent_id")  # 对应 agent_id
-        self.api_key = config.get("api_key")
-        self.base_url = config.get("base_url", config.get("url"))  # 默认使用 base_url
-        self.api_url = f"{self.base_url}/api/conversation/process"  # 拼接完整的 API URL
-
-    def response(self, session_id, dialogue):
-        print(dialogue)
-        try:
-            # home assistant语音助手自带意图，无需使用xiaozhi ai自带的，只需要把用户说的话传递给home assistant即可
-
-            # 提取最后一个 role 为 'user' 的 content
-            input_text = None
-            if isinstance(dialogue, list):  # 确保 dialogue 是一个列表
-                # 逆序遍历，找到最后一个 role 为 'user' 的消息
-                for message in reversed(dialogue):
-                    if message.get("role") == "user":  # 找到 role 为 'user' 的消息
-                        input_text = message.get("content", "")
-                        break  # 找到后立即退出循环
-
-            # 构造请求数据
-            payload = {
-                "text": input_text,
-                "agent_id": self.agent_id,
-                "conversation_id": session_id  # 使用 session_id 作为 conversation_id
-            }
-            # 设置请求头
-            headers = {
-                "Authorization": f"Bearer {self.api_key}",
-                "Content-Type": "application/json"
-            }
-
-            # 发起 POST 请求
-            response = requests.post(self.api_url, json=payload, headers=headers)
-
-            # 检查请求是否成功
-            response.raise_for_status()
-
-            # 解析返回数据
-            data = response.json()
-            speech = data.get("response", {}).get("speech", {}).get("plain", {}).get("speech", "")
-
-            # 返回生成的内容
-            if speech:
-                yield speech
-            else:
-                logger.bind(tag=TAG).warning("API 返回数据中没有 speech 内容")
-
-        except RequestException as e:
-            logger.bind(tag=TAG).error(f"HTTP 请求错误: {e}")
-        except Exception as e:
-            logger.bind(tag=TAG).error(f"生成响应时出错: {e}")

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

@@ -64,5 +64,5 @@ class LLMProvider(LLMProviderBase):
                 yield chunk.choices[0].delta.content, chunk.choices[0].delta.tool_calls
 
         except Exception as e:
-            self.logger.bind(tag=TAG).error(f"Error in function call streaming: {e}")
+            logger.bind(tag=TAG).error(f"Error in function call streaming: {e}")
             yield {"type": "content", "content": f"【OpenAI服务响应异常: {e}】"}

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

@@ -0,0 +1,85 @@
+from config.logger import setup_logging
+from openai import OpenAI
+import json
+from core.providers.llm.base import LLMProviderBase
+
+TAG = __name__
+logger = setup_logging()
+
+
+class LLMProvider(LLMProviderBase):
+    def __init__(self, config):
+        self.model_name = config.get("model_name")
+        self.base_url = config.get("base_url", "http://localhost:9997")
+        # Initialize OpenAI client with Xinference base URL
+        # 如果没有v1，增加v1
+        if not self.base_url.endswith("/v1"):
+            self.base_url = f"{self.base_url}/v1"
+            
+        logger.bind(tag=TAG).info(f"Initializing Xinference LLM provider with model: {self.model_name}, base_url: {self.base_url}")
+
+        try:
+            self.client = OpenAI(
+                base_url=self.base_url,
+                api_key="xinference"  # Xinference has a similar setup to Ollama where it doesn't need an actual key
+            )
+            logger.bind(tag=TAG).info("Xinference client initialized successfully")
+        except Exception as e:
+            logger.bind(tag=TAG).error(f"Error initializing Xinference client: {e}")
+            raise
+
+    def response(self, session_id, dialogue):
+        try:
+            logger.bind(tag=TAG).debug(f"Sending request to Xinference with model: {self.model_name}, dialogue length: {len(dialogue)}")
+            responses = self.client.chat.completions.create(
+                model=self.model_name,
+                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:
+                        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}")
+
+        except Exception as e:
+            logger.bind(tag=TAG).error(f"Error in Xinference response generation: {e}")
+            yield "【Xinference服务响应异常】"
+
+    def response_with_functions(self, session_id, dialogue, functions=None):
+        try:
+            logger.bind(tag=TAG).debug(f"Sending function call request to Xinference with model: {self.model_name}, dialogue length: {len(dialogue)}")
+            if functions:
+                logger.bind(tag=TAG).debug(f"Function calls enabled with: {[f.get('function', {}).get('name') for f in functions]}")
+                
+            stream = self.client.chat.completions.create(
+                model=self.model_name,
+                messages=dialogue,
+                stream=True,
+                tools=functions,
+            )
+
+            for chunk in stream:
+                delta = chunk.choices[0].delta
+                content = delta.content
+                tool_calls = delta.tool_calls
+                
+                if content:
+                    yield content, tool_calls
+                elif tool_calls:
+                    yield None, tool_calls
+
+        except Exception as e:
+            logger.bind(tag=TAG).error(f"Error in Xinference function call: {e}")
+            yield {"type": "content", "content": f"【Xinference服务响应异常: {str(e)}】"} 

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

@@ -121,9 +121,15 @@ class TTSProviderBase(ABC):
     async def tts_one_sentence(self, text, u_id=None):
         if not u_id:
             u_id = str(uuid.uuid4()).replace("-", "")
-        self.tts.tts_text_queue.put(TTSMessageDTO(u_id=u_id, msg_type=MsgType.START_TTS_REQUEST, content=''))
+        self.tts.tts_text_queue.put(
-        self.tts.tts_text_queue.put(TTSMessageDTO(u_id=u_id, msg_type=MsgType.TTS_TEXT_REQUEST, content=text))
+            TTSMessageDTO(u_id=u_id, msg_type=MsgType.START_TTS_REQUEST, content="")
-        self.tts.tts_text_queue.put(TTSMessageDTO(u_id=u_id, msg_type=MsgType.STOP_TTS_REQUEST, content=text))
+        )
+        self.tts.tts_text_queue.put(
+            TTSMessageDTO(u_id=u_id, msg_type=MsgType.TTS_TEXT_REQUEST, content=text)
+        )
+        self.tts.tts_text_queue.put(
+            TTSMessageDTO(u_id=u_id, msg_type=MsgType.STOP_TTS_REQUEST, content=text)
+        )
 
     def _enable_two_way_tts(self):
         while not self.stop_event.is_set():
@@ -273,7 +279,6 @@ class TTSProviderBase(ABC):
     def generate_filename(self):
         pass
 
-
     @abstractmethod
     async def text_to_speak(self, u_id, text, is_last_text=False, is_first_text=False):
         pass

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

@@ -9,7 +9,6 @@ import json
 import base64
 import requests
 from datetime import datetime
-from mutagen.oggopus import OggOpus
 
 import websockets
 
@@ -76,14 +75,16 @@ EVENT_TTSResponse = 352
 
 
 class Header:
-    def __init__(self,
+    def __init__(
-                 protocol_version=PROTOCOL_VERSION,
+        self,
-                 header_size=DEFAULT_HEADER_SIZE,
+        protocol_version=PROTOCOL_VERSION,
-                 message_type: int = 0,
+        header_size=DEFAULT_HEADER_SIZE,
-                 message_type_specific_flags: int = 0,
+        message_type: int = 0,
-                 serial_method: int = NO_SERIALIZATION,
+        message_type_specific_flags: int = 0,
-                 compression_type: int = COMPRESSION_NO,
+        serial_method: int = NO_SERIALIZATION,
-                 reserved_data=0):
+        compression_type: int = COMPRESSION_NO,
+        reserved_data=0,
+    ):
         self.header_size = header_size
         self.protocol_version = protocol_version
         self.message_type = message_type
@@ -93,16 +94,20 @@ class Header:
         self.reserved_data = reserved_data
 
     def as_bytes(self) -> bytes:
-        return bytes([
+        return bytes(
-            (self.protocol_version << 4) | self.header_size,
+            [
-            (self.message_type << 4) | self.message_type_specific_flags,
+                (self.protocol_version << 4) | self.header_size,
-            (self.serial_method << 4) | self.compression_type,
+                (self.message_type << 4) | self.message_type_specific_flags,
-            self.reserved_data
+                (self.serial_method << 4) | self.compression_type,
-        ])
+                self.reserved_data,
+            ]
+        )
 
 
 class Optional:
-    def __init__(self, event: int = EVENT_NONE, sessionId: str = None, sequence: int = None):
+    def __init__(
+        self, event: int = EVENT_NONE, sessionId: str = None, sequence: int = None
+    ):
         self.event = event
         self.sessionId = sessionId
         self.errorCode: int = 0
@@ -160,45 +165,57 @@ class TTSProvider(TTSProviderBase):
             "X-Api-Resource-Id": self.resource_id,
             "X-Api-Connect-Id": uuid.uuid4(),
         }
-        self.ws = await websockets.connect(self.ws_url, additional_headers=ws_header, max_size=1000000000)
+        self.ws = await websockets.connect(
-        tts_priority = threading.Thread(target=self._start_monitor_tts_response_thread(), daemon=True)
+            self.ws_url, additional_headers=ws_header, max_size=1000000000
+        )
+        tts_priority = threading.Thread(
+            target=self._start_monitor_tts_response_thread(), daemon=True
+        )
         tts_priority.start()
 
     def generate_filename(self, extension=".wav"):
-        return os.path.join(self.output_file, f"tts-{datetime.now().date()}@{uuid.uuid4().hex}{extension}")
+        return os.path.join(
+            self.output_file,
+            f"tts-{datetime.now().date()}@{uuid.uuid4().hex}{extension}",
+        )
 
-    async def send_event(self, header: bytes, optional: bytes | None = None,
+    async def send_event(
-                         payload: bytes = None):
+        self, header: bytes, optional: bytes | None = None, payload: bytes = None
+    ):
         full_client_request = bytearray(header)
         if optional is not None:
             full_client_request.extend(optional)
         if payload is not None:
-            payload_size = len(payload).to_bytes(4, 'big', signed=True)
+            payload_size = len(payload).to_bytes(4, "big", signed=True)
             full_client_request.extend(payload_size)
             full_client_request.extend(payload)
         await self.ws.send(full_client_request)
 
     async def send_text(self, speaker: str, text: str, session_id):
-        header = Header(message_type=FULL_CLIENT_REQUEST,
+        header = Header(
-                        message_type_specific_flags=MsgTypeFlagWithEvent,
+            message_type=FULL_CLIENT_REQUEST,
-                        serial_method=JSON).as_bytes()
+            message_type_specific_flags=MsgTypeFlagWithEvent,
+            serial_method=JSON,
+        ).as_bytes()
         optional = Optional(event=EVENT_TaskRequest, sessionId=session_id).as_bytes()
-        payload = self.get_payload_bytes(event=EVENT_TaskRequest, text=text, speaker=speaker)
+        payload = self.get_payload_bytes(
+            event=EVENT_TaskRequest, text=text, speaker=speaker
+        )
         return await self.send_event(header, optional, payload)
 
     # 读取 res 数组某段 字符串内容
     def read_res_content(self, res: bytes, offset: int):
-        content_size = int.from_bytes(res[offset: offset + 4], "big", signed=True)
+        content_size = int.from_bytes(res[offset : offset + 4], "big", signed=True)
         offset += 4
-        content = str(res[offset: offset + content_size])
+        content = str(res[offset : offset + content_size])
         offset += content_size
         return content, offset
 
     # 读取 payload
     def read_res_payload(self, res: bytes, offset: int):
-        payload_size = int.from_bytes(res[offset: offset + 4], "big", signed=True)
+        payload_size = int.from_bytes(res[offset : offset + 4], "big", signed=True)
         offset += 4
-        payload = res[offset: offset + payload_size]
+        payload = res[offset : offset + payload_size]
         offset += payload_size
         return payload, offset
 
@@ -211,11 +228,11 @@ class TTSProvider(TTSProviderBase):
         header = response.header
         num = 0b00001111
         header.protocol_version = res[0] >> 4 & num
-        header.header_size = res[0] & 0x0f
+        header.header_size = res[0] & 0x0F
         header.message_type = (res[1] >> 4) & num
-        header.message_type_specific_flags = res[1] & 0x0f
+        header.message_type_specific_flags = res[1] & 0x0F
         header.serialization_method = res[2] >> num
-        header.message_compression = res[2] & 0x0f
+        header.message_compression = res[2] & 0x0F
         header.reserved = res[3]
         #
         offset = 4
@@ -231,78 +248,101 @@ class TTSProvider(TTSProviderBase):
                 elif optional.event == EVENT_ConnectionStarted:
                     optional.connectionId, offset = self.read_res_content(res, offset)
                 elif optional.event == EVENT_ConnectionFailed:
-                    optional.response_meta_json, offset = self.read_res_content(res, offset)
+                    optional.response_meta_json, offset = self.read_res_content(
-                elif (optional.event == EVENT_SessionStarted
+                        res, offset
-                      or optional.event == EVENT_SessionFailed
+                    )
-                      or optional.event == EVENT_SessionFinished):
+                elif (
+                    optional.event == EVENT_SessionStarted
+                    or optional.event == EVENT_SessionFailed
+                    or optional.event == EVENT_SessionFinished
+                ):
                     optional.sessionId, offset = self.read_res_content(res, offset)
-                    optional.response_meta_json, offset = self.read_res_content(res, offset)
+                    optional.response_meta_json, offset = self.read_res_content(
+                        res, offset
+                    )
                 else:
                     optional.sessionId, offset = self.read_res_content(res, offset)
                     response.payload, offset = self.read_res_payload(res, offset)
 
         elif header.message_type == ERROR_INFORMATION:
-            optional.errorCode = int.from_bytes(res[offset:offset + 4], "big", signed=True)
+            optional.errorCode = int.from_bytes(
+                res[offset : offset + 4], "big", signed=True
+            )
             offset += 4
             response.payload, offset = self.read_res_payload(res, offset)
         return response
 
     async def start_connection(self):
-        header = Header(message_type=FULL_CLIENT_REQUEST, message_type_specific_flags=MsgTypeFlagWithEvent).as_bytes()
+        header = Header(
+            message_type=FULL_CLIENT_REQUEST,
+            message_type_specific_flags=MsgTypeFlagWithEvent,
+        ).as_bytes()
         optional = Optional(event=EVENT_Start_Connection).as_bytes()
         payload = str.encode("{}")
         return await self.send_event(header, optional, payload)
 
     def print_response(self, res, tag_msg: str):
-        logger.bind(tag=TAG).info(f'===>{tag_msg} header:{res.header.__dict__}')
+        logger.bind(tag=TAG).info(f"===>{tag_msg} header:{res.header.__dict__}")
-        logger.bind(tag=TAG).info(f'===>{tag_msg} optional:{res.optional.__dict__}')
+        logger.bind(tag=TAG).info(f"===>{tag_msg} optional:{res.optional.__dict__}")
 
-    def get_payload_bytes(self, uid='1234', event=EVENT_NONE, text='', speaker='', audio_format='pcm',
+    def get_payload_bytes(
-                          audio_sample_rate=16000):
+        self,
-        return str.encode(json.dumps(
+        uid="1234",
-            {
+        event=EVENT_NONE,
-                "user": {"uid": uid},
+        text="",
-                "event": event,
+        speaker="",
-                "namespace": "BidirectionalTTS",
+        audio_format="pcm",
-                "req_params": {
+        audio_sample_rate=16000,
-                    "text": text,
+    ):
-                    "speaker": speaker,
+        return str.encode(
-                    "audio_params": {
+            json.dumps(
-                        "format": audio_format,
+                {
-                        "sample_rate": audio_sample_rate
+                    "user": {"uid": uid},
-                    }
+                    "event": event,
+                    "namespace": "BidirectionalTTS",
+                    "req_params": {
+                        "text": text,
+                        "speaker": speaker,
+                        "audio_params": {
+                            "format": audio_format,
+                            "sample_rate": audio_sample_rate,
+                        },
+                    },
                 }
-            }
+            )
-        ))
+        )
 
     async def finish_connection(self):
-        header = Header(message_type=FULL_CLIENT_REQUEST,
+        header = Header(
-                        message_type_specific_flags=MsgTypeFlagWithEvent,
+            message_type=FULL_CLIENT_REQUEST,
-                        serial_method=JSON
+            message_type_specific_flags=MsgTypeFlagWithEvent,
-                        ).as_bytes()
+            serial_method=JSON,
+        ).as_bytes()
         optional = Optional(event=EVENT_FinishConnection).as_bytes()
-        payload = str.encode('{}')
+        payload = str.encode("{}")
         await self.send_event(header, optional, payload)
         return
 
     async def start_session(self, session_id):
         self.stop_event_response.clear()
-        header = Header(message_type=FULL_CLIENT_REQUEST,
+        header = Header(
-                        message_type_specific_flags=MsgTypeFlagWithEvent,
+            message_type=FULL_CLIENT_REQUEST,
-                        serial_method=JSON
+            message_type_specific_flags=MsgTypeFlagWithEvent,
-                        ).as_bytes()
+            serial_method=JSON,
+        ).as_bytes()
         optional = Optional(event=EVENT_StartSession, sessionId=session_id).as_bytes()
         payload = self.get_payload_bytes(event=EVENT_StartSession, speaker=self.speaker)
         await self.send_event(header, optional, payload)
 
     async def finish_session(self, session_id):
         self.stop_event_response.set()
-        header = Header(message_type=FULL_CLIENT_REQUEST,
+        header = Header(
-                        message_type_specific_flags=MsgTypeFlagWithEvent,
+            message_type=FULL_CLIENT_REQUEST,
-                        serial_method=JSON
+            message_type_specific_flags=MsgTypeFlagWithEvent,
-                        ).as_bytes()
+            serial_method=JSON,
+        ).as_bytes()
         optional = Optional(event=EVENT_FinishSession, sessionId=session_id).as_bytes()
-        payload = str.encode('{}')
+        payload = str.encode("{}")
         await self.send_event(header, optional, payload)
         return
 
@@ -327,59 +367,78 @@ class TTSProvider(TTSProviderBase):
 
     def _start_monitor_tts_response_thread(self):
         # 初始化链接
-        asyncio.run_coroutine_threadsafe(self._start_monitor_tts_response(), loop=self.loop)
+        asyncio.run_coroutine_threadsafe(
+            self._start_monitor_tts_response(), loop=self.loop
+        )
 
     async def _start_monitor_tts_response(self):
-        chunk_total = b''
+        chunk_total = b""
         while True:
             try:
                 msg = await self.ws.recv()  # 确保 `recv()` 运行在同一个 event loop
                 res = self.parser_response(msg)
-                self.print_response(res, 'send_text res:')
+                self.print_response(res, "send_text res:")
 
-                if res.optional.event == EVENT_TTSResponse and res.header.message_type == AUDIO_ONLY_RESPONSE:
+                if (
-                    logger.bind(tag=TAG).info(f'推送数据到队列里面～～')
+                    res.optional.event == EVENT_TTSResponse
+                    and res.header.message_type == AUDIO_ONLY_RESPONSE
+                ):
+                    logger.bind(tag=TAG).info(f"推送数据到队列里面～～")
                     opus_datas = self.wav_to_opus_data_audio_raw(res.payload)
                     self.tts_audio_queue.put(
                         TTSMessageDTO(
-                            u_id=self.u_id, msg_type=MsgType.TTS_TEXT_RESPONSE, content=opus_datas,
+                            u_id=self.u_id,
-                            tts_finish_text="", sentence_type=None, duration=0
+                            msg_type=MsgType.TTS_TEXT_RESPONSE,
+                            content=opus_datas,
+                            tts_finish_text="",
+                            sentence_type=None,
+                            duration=0,
                         )
                     )
                 elif res.optional.event == EVENT_TTSSentenceStart:
-                    json_data = json.loads(res.payload.decode('utf-8'))
+                    json_data = json.loads(res.payload.decode("utf-8"))
                     self.tts_text = json_data.get("text", "")
-                    logger.bind(tag=TAG).info(f'句子开始～～{self.tts_text}')
+                    logger.bind(tag=TAG).info(f"句子开始～～{self.tts_text}")
                     self.tts_audio_queue.put(
                         TTSMessageDTO(
-                            u_id=self.u_id, msg_type=MsgType.TTS_TEXT_RESPONSE, content=[],
+                            u_id=self.u_id,
+                            msg_type=MsgType.TTS_TEXT_RESPONSE,
+                            content=[],
                             tts_finish_text=self.tts_text,
-                            sentence_type=SentenceType.SENTENCE_START
+                            sentence_type=SentenceType.SENTENCE_START,
                         )
                     )
                 elif res.optional.event == EVENT_TTSSentenceEnd:
-                    logger.bind(tag=TAG).info(f'句子结束～～{self.tts_text}')
+                    logger.bind(tag=TAG).info(f"句子结束～～{self.tts_text}")
                     self.tts_audio_queue.put(
                         TTSMessageDTO(
-                            u_id=self.u_id, msg_type=MsgType.TTS_TEXT_RESPONSE, content=[],
+                            u_id=self.u_id,
+                            msg_type=MsgType.TTS_TEXT_RESPONSE,
+                            content=[],
                             tts_finish_text=self.tts_text,
-                            sentence_type=SentenceType.SENTENCE_END
+                            sentence_type=SentenceType.SENTENCE_END,
                         )
                     )
                 elif res.optional.event == EVENT_SessionFinished:
-                    logger.bind(tag=TAG).info(f'会话结束～～,最后一句补零')
+                    logger.bind(tag=TAG).info(f"会话结束～～,最后一句补零")
-                    opus_datas = self.wav_to_opus_data_audio_raw(b'', is_end=True)
+                    opus_datas = self.wav_to_opus_data_audio_raw(b"", is_end=True)
                     self.tts_audio_queue.put(
                         TTSMessageDTO(
-                            u_id=self.u_id, msg_type=MsgType.TTS_TEXT_RESPONSE, content=opus_datas,
+                            u_id=self.u_id,
-                            tts_finish_text="", sentence_type=None, duration=0
+                            msg_type=MsgType.TTS_TEXT_RESPONSE,
+                            content=opus_datas,
+                            tts_finish_text="",
+                            sentence_type=None,
+                            duration=0,
                         )
                     )
                     self.tts_audio_queue.put(
                         TTSMessageDTO(
-                            u_id=self.u_id, msg_type=MsgType.STOP_TTS_RESPONSE, content=[],
+                            u_id=self.u_id,
+                            msg_type=MsgType.STOP_TTS_RESPONSE,
+                            content=[],
                             tts_finish_text=self.tts_text,
-                            sentence_type=SentenceType.SENTENCE_END
+                            sentence_type=SentenceType.SENTENCE_END,
                         )
                     )
                 else:

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

@@ -0,0 +1,156 @@
+import hashlib
+import hmac
+import os
+import time
+import uuid
+import json
+import base64
+import requests
+from datetime import datetime, timezone
+from core.providers.tts.base import TTSProviderBase
+
+
+class TTSProvider(TTSProviderBase):
+    def __init__(self, config, delete_audio_file):
+        super().__init__(config, delete_audio_file)
+        self.appid = config.get("appid")
+        self.secret_id = config.get("secret_id")
+        self.secret_key = config.get("secret_key")
+        self.voice = config.get("voice")
+        self.api_url = "https://tts.tencentcloudapi.com"  # 正确的API端点
+        self.region = config.get("region")
+        self.output_file = config.get("output_dir")
+
+    def _get_auth_headers(self, request_body):
+        """生成鉴权请求头"""
+        # 获取当前UTC时间戳
+        timestamp = int(time.time())
+        
+        # 使用UTC时间计算日期
+        utc_date = datetime.fromtimestamp(timestamp, tz=timezone.utc).strftime('%Y-%m-%d')
+        
+        # 服务名称必须是 "tts"
+        service = "tts"
+        
+        # 拼接凭证范围
+        credential_scope = f"{utc_date}/{service}/tc3_request"
+        
+        # 使用TC3-HMAC-SHA256签名方法
+        algorithm = "TC3-HMAC-SHA256"
+        
+        # 构建规范请求字符串
+        http_request_method = "POST"
+        canonical_uri = "/"
+        canonical_querystring = ""
+        
+        # 请求头必须包含host和content-type，且按字典序排列
+        canonical_headers = (
+            f"content-type:application/json\n"
+            f"host:tts.tencentcloudapi.com\n"
+        )
+        signed_headers = "content-type;host"
+        
+        # 请求体哈希值
+        payload = json.dumps(request_body)
+        payload_hash = hashlib.sha256(payload.encode('utf-8')).hexdigest()
+        
+        # 构建规范请求字符串
+        canonical_request = (
+            f"{http_request_method}\n"
+            f"{canonical_uri}\n"
+            f"{canonical_querystring}\n"
+            f"{canonical_headers}\n"
+            f"{signed_headers}\n"
+            f"{payload_hash}"
+        )
+        
+        # 计算规范请求的哈希值
+        hashed_canonical_request = hashlib.sha256(canonical_request.encode('utf-8')).hexdigest()
+        
+        # 构建待签名字符串
+        string_to_sign = (
+            f"{algorithm}\n"
+            f"{timestamp}\n"
+            f"{credential_scope}\n"
+            f"{hashed_canonical_request}"
+        )
+        
+        # 计算签名密钥
+        secret_date = self._hmac_sha256(f"TC3{self.secret_key}".encode('utf-8'), utc_date)
+        secret_service = self._hmac_sha256(secret_date, service)
+        secret_signing = self._hmac_sha256(secret_service, "tc3_request")
+        
+        # 计算签名
+        signature = hmac.new(
+            secret_signing, 
+            string_to_sign.encode('utf-8'), 
+            hashlib.sha256
+        ).hexdigest()
+        
+        # 构建授权头
+        authorization = (
+            f"{algorithm} "
+            f"Credential={self.secret_id}/{credential_scope}, "
+            f"SignedHeaders={signed_headers}, "
+            f"Signature={signature}"
+        )
+        
+        # 构建请求头
+        headers = {
+            "Content-Type": "application/json",
+            "Host": "tts.tencentcloudapi.com",
+            "Authorization": authorization,
+            "X-TC-Action": "TextToVoice",
+            "X-TC-Timestamp": str(timestamp),
+            "X-TC-Version": "2019-08-23",
+            "X-TC-Region": self.region,
+            "X-TC-Language": "zh-CN"
+        }
+        
+        return headers
+    
+    def _hmac_sha256(self, key, msg):
+        """HMAC-SHA256加密"""
+        if isinstance(msg, str):
+            msg = msg.encode('utf-8')
+        return hmac.new(key, msg, hashlib.sha256).digest()
+
+    def generate_filename(self, extension=".wav"):
+        return os.path.join(self.output_file, f"tts-{datetime.now().date()}@{uuid.uuid4().hex}{extension}")
+
+    async def text_to_speak(self, text, output_file):
+        # 构建请求体
+        request_json = {
+            "Text": text,  # 合成语音的源文本
+            "SessionId": str(uuid.uuid4()),  # 会话ID，随机生成
+            "VoiceType": int(self.voice),  # 音色
+        }
+
+        try:
+            # 获取请求头（每次请求都重新生成，以确保时间戳和签名是最新的）
+            headers = self._get_auth_headers(request_json)
+            
+            # 发送请求
+            resp = requests.post(self.api_url, json.dumps(request_json), headers=headers)
+            
+            # 检查响应
+            if resp.status_code == 200:
+                response_data = resp.json()
+                
+                # 检查是否成功
+                if response_data.get("Response", {}).get("Error") is not None:
+                    error_info = response_data["Response"]["Error"]
+                    raise Exception(f"API返回错误: {error_info['Code']}: {error_info['Message']}")
+                
+                # 提取音频数据
+                audio_data = response_data["Response"].get("Audio")
+                if audio_data:
+                    # 解码Base64音频数据并保存
+                    with open(output_file, "wb") as f:
+                        f.write(base64.b64decode(audio_data))
+                else:
+                    raise Exception(f"{__name__}: 没有返回音频数据: {response_data}")
+            else:
+                raise Exception(f"{__name__} status_code: {resp.status_code} response: {resp.content}")
+        except Exception as e:
+            raise Exception(f"{__name__} error: {e}")

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

@@ -35,6 +35,15 @@ class Dialogue:
             self.getMessages(m, dialogue)
         return dialogue
 
+    def update_system_message(self, new_content: str):
+        """更新或添加系统消息"""
+        # 查找第一个系统消息
+        system_msg = next((msg for msg in self.dialogue if msg.role == "system"), None)
+        if system_msg:
+            system_msg.content = new_content
+        else:
+            self.put(Message(role="system", content=new_content))
+
     def get_llm_dialogue_with_memory(self, memory_str: str = None) -> List[Dict[str, str]]:
         if memory_str is None or len(memory_str) == 0:
             return self.get_llm_dialogue()

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

@@ -1,4 +1,5 @@
 import os
+import re
 import sys
 from config.logger import setup_logging
 import importlib
@@ -14,4 +15,98 @@ def create_instance(class_name, *args, **kwargs):
             sys.modules[lib_name] = importlib.import_module(f'{lib_name}')
         return sys.modules[lib_name].TTSProvider(*args, **kwargs)
 
-    raise ValueError(f"不支持的TTS类型: {class_name}，请检查该配置的type是否设置正确")
+    raise ValueError(f"不支持的TTS类型: {class_name}，请检查该配置的type是否设置正确")
+
+
+class MarkdownCleaner:
+    """
+    封装 Markdown 清理逻辑：直接用 MarkdownCleaner.clean_markdown(text) 即可
+    """
+    # 公式字符
+    NORMAL_FORMULA_CHARS = re.compile(r'[a-zA-Z\\^_{}\+\-\(\)\[\]=]')
+
+    @staticmethod
+    def _replace_inline_dollar(m: re.Match) -> str:
+        """
+        只要捕获到完整的 "$...$":
+          - 如果内部有典型公式字符 => 去掉两侧 $
+          - 否则 (纯数字/货币等) => 保留 "$...$"
+        """
+        content = m.group(1)
+        if MarkdownCleaner.NORMAL_FORMULA_CHARS.search(content):
+            return content
+        else:
+            return m.group(0)
+
+    @staticmethod
+    def _replace_table_block(match: re.Match) -> str:
+        """
+        当匹配到一个整段表格块时，回调该函数。
+        """
+        block_text = match.group('table_block')
+        lines = block_text.strip('\n').split('\n')
+
+        parsed_table = []
+        for line in lines:
+            line_stripped = line.strip()
+            if re.match(r'^\|\s*[-:]+\s*(\|\s*[-:]+\s*)+\|?$', line_stripped):
+                continue
+            columns = [col.strip() for col in line_stripped.split('|') if col.strip() != '']
+            if columns:
+                parsed_table.append(columns)
+
+        if not parsed_table:
+            return ""
+
+        headers = parsed_table[0]
+        data_rows = parsed_table[1:] if len(parsed_table) > 1 else []
+
+        lines_for_tts = []
+        if len(parsed_table) == 1:
+            # 只有一行
+            only_line_str = ", ".join(parsed_table[0])
+            lines_for_tts.append(f"单行表格：{only_line_str}")
+        else:
+            lines_for_tts.append(f"表头是：{', '.join(headers)}")
+            for i, row in enumerate(data_rows, start=1):
+                row_str_list = []
+                for col_index, cell_val in enumerate(row):
+                    if col_index < len(headers):
+                        row_str_list.append(f"{headers[col_index]} = {cell_val}")
+                    else:
+                        row_str_list.append(cell_val)
+                lines_for_tts.append(f"第 {i} 行：{', '.join(row_str_list)}")
+
+        return "\n".join(lines_for_tts) + "\n"
+
+    # 预编译所有正则表达式（按执行频率排序）
+    # 这里要把 replace_xxx 的静态方法放在最前定义，以便在列表里能正确引用它们。
+    REGEXES = [
+        (re.compile(r'```.*?```', re.DOTALL), ''),  # 代码块
+        (re.compile(r'^#+\s*', re.MULTILINE), ''),  # 标题
+        (re.compile(r'(\*\*|__)(.*?)\1'), r'\2'),  # 粗体
+        (re.compile(r'(\*|_)(?=\S)(.*?)(?<=\S)\1'), r'\2'),  # 斜体
+        (re.compile(r'!\[.*?\]\(.*?\)'), ''),  # 图片
+        (re.compile(r'\[(.*?)\]\(.*?\)'), r'\1'),  # 链接
+        (re.compile(r'^\s*>+\s*', re.MULTILINE), ''),  # 引用
+        (
+            re.compile(r'(?P<table_block>(?:^[^\n]*\|[^\n]*\n)+)', re.MULTILINE),
+            _replace_table_block
+        ),
+        (re.compile(r'^\s*[*+-]\s*', re.MULTILINE), '- '),  # 列表
+        (re.compile(r'\$\$.*?\$\$', re.DOTALL), ''),  # 块级公式
+        (
+            re.compile(r'(?<![A-Za-z0-9])\$([^\n$]+)\$(?![A-Za-z0-9])'),
+            _replace_inline_dollar
+        ),
+        (re.compile(r'\n{2,}'), '\n'),  # 多余空行
+    ]
+
+    @staticmethod
+    def clean_markdown(text: str) -> str:
+        """
+        主入口方法：依序执行所有正则，移除或替换 Markdown 元素
+        """
+        for regex, replacement in MarkdownCleaner.REGEXES:
+            text = regex.sub(replacement, text)
+        return text.strip()

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

@@ -67,15 +67,11 @@ def is_private_ip(ip_addr):
 
 def get_ip_info(ip_addr):
     try:
-        base_url = "https://freeipapi.com/api/json"
+        url = "https://whois.pconline.com.cn/ipJson.jsp?json=true"
-        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"),
+            "city": resp.get("city")
-            "region": resp.get("regionName"), 
-            "country": resp.get("countryName")
         }
         return ip_info
     except Exception as e:

main/xiaozhi-server/core/websocket_server.py

@@ -85,24 +85,25 @@ class WebSocketServer:
         server_config = self.config["server"]
         host = server_config["ip"]
         port = server_config["port"]
-        selected_module = self.config.get("selected_module")
-        self.logger.bind(tag=TAG).info(
-            f"selected_module values: {', '.join(selected_module.values())}"
-        )
 
         self.logger.bind(tag=TAG).info(
-            "Server is running at ws://{}:{}", get_local_ip(), port
+            "Server is running at ws://{}:{}/xiaozhi/v1/", get_local_ip(), port
         )
         self.logger.bind(tag=TAG).info(
             "=======上面的地址是websocket协议地址，请勿用浏览器访问======="
         )
+        self.logger.bind(tag=TAG).info(
+            "如想测试websocket请用谷歌浏览器打开test目录下的test_page.html"
+        )
+        self.logger.bind(tag=TAG).info(
+            "=============================================================\n"
+        )
         async with websockets.serve(self._handle_connection, host, port):
             await asyncio.Future()
 
     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
-        # tts 变成链接的时候创建，避免并非问题
         handler = ConnectionHandler(
             self.config,
             self._vad,

main/xiaozhi-server/mcp_server_settings.json

@@ -0,0 +1,31 @@
+{
+  "des": [
+    "在data目录下创建.mcp_server_settings.json文件，可以选择下面的MCP服务，也可以自行添加新的MCP服务。",
+    "后面不断测试补充好用的mcp服务，欢迎大家一起补充。",
+    "记得删除注释行,des属性仅为说明,不会被解析。",
+    "des和link属性，仅为说明安装方式，方便大家查看原始链接，不是必须项。"
+  ],
+  "mcpServers": {
+    "filesystem": {
+      "command": "npx",
+      "args": [
+        "-y",
+        "@modelcontextprotocol/server-filesystem",
+        "/Users/username/Desktop",
+        "/path/to/other/allowed/dir"
+      ],
+      "link":"https://github.com/modelcontextprotocol/servers/tree/main/src/filesystem"
+    },
+    "playwright": {
+      "command": "npx",
+      "args": ["-y", "@executeautomation/playwright-mcp-server"],
+      "des" : "run 'npx playwright install' first",
+      "link": "https://github.com/executeautomation/mcp-playwright"
+    },
+    "windows-cli": {
+      "command": "npx",
+      "args": ["-y", "@simonb97/server-win-cli"],
+      "link": "https://github.com/SimonB97/win-cli-mcp-server"
+    }
+  }
+}

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

@@ -0,0 +1,206 @@
+import random
+import requests
+import xml.etree.ElementTree as ET
+from bs4 import BeautifulSoup
+from config.logger import setup_logging
+from plugins_func.register import register_function, ToolType, ActionResponse, Action
+
+TAG = __name__
+logger = setup_logging()
+
+GET_NEWS_FUNCTION_DESC = {
+    "type": "function",
+    "function": {
+        "name": "get_news",
+        "description": (
+            "获取最新新闻，随机选择一条新闻进行播报。"
+            "用户可以指定新闻类型，如社会新闻、科技新闻、国际新闻等。"
+            "如果没有指定，默认播报社会新闻。"
+            "用户可以要求获取详细内容，此时会获取新闻的详细内容。"
+        ),
+        "parameters": {
+            "type": "object",
+            "properties": {
+                "category": {
+                    "type": "string",
+                    "description": "新闻类别，例如社会、科技、国际。可选参数，如果不提供则使用默认类别"
+                },
+                "detail": {
+                    "type": "boolean",
+                    "description": "是否获取详细内容，默认为false。如果为true，则获取上一条新闻的详细内容"
+                },
+                "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": ["lang"]
+        }
+    }
+}
+
+
+def fetch_news_from_rss(rss_url):
+    """从RSS源获取新闻列表"""
+    try:
+        response = requests.get(rss_url)
+        response.raise_for_status()
+
+        # 解析XML
+        root = ET.fromstring(response.content)
+
+        # 查找所有item元素（新闻条目）
+        news_items = []
+        for item in root.findall('.//item'):
+            title = item.find('title').text if item.find('title') is not None else "无标题"
+            link = item.find('link').text if item.find('link') is not None else "#"
+            description = item.find('description').text if item.find('description') is not None else "无描述"
+            pubDate = item.find('pubDate').text if item.find('pubDate') is not None else "未知时间"
+
+            news_items.append({
+                'title': title,
+                'link': link,
+                'description': description,
+                'pubDate': pubDate
+            })
+
+        return news_items
+    except Exception as e:
+        logger.bind(tag=TAG).error(f"获取RSS新闻失败: {e}")
+        return []
+
+
+def fetch_news_detail(url):
+    """获取新闻详情页内容并总结"""
+    try:
+        response = requests.get(url)
+        response.raise_for_status()
+
+        soup = BeautifulSoup(response.content, 'html.parser')
+
+        # 尝试提取正文内容 (这里的选择器需要根据实际网站结构调整)
+        content_div = soup.select_one('.content_desc, .content, article, .article-content')
+        if content_div:
+            paragraphs = content_div.find_all('p')
+            content = '\n'.join([p.get_text().strip() for p in paragraphs if p.get_text().strip()])
+            return content
+        else:
+            # 如果找不到特定的内容区域，尝试获取所有段落
+            paragraphs = soup.find_all('p')
+            content = '\n'.join([p.get_text().strip() for p in paragraphs if p.get_text().strip()])
+            return content[:2000]  # 限制长度
+    except Exception as e:
+        logger.bind(tag=TAG).error(f"获取新闻详情失败: {e}")
+        return "无法获取详细内容"
+
+
+def map_category(category_text):
+    """将用户输入的中文类别映射到配置文件中的类别键"""
+    if not category_text:
+        return None
+
+    # 类别映射字典，目前支持社会、国际、财经新闻，如需更多类型，参见配置文件
+    category_map = {
+        # 社会新闻
+        "社会": "society",
+        "社会新闻": "society",
+        # 国际新闻
+        "国际": "world",
+        "国际新闻": "world",
+        # 财经新闻
+        "财经": "finance",
+        "财经新闻": "finance",
+        "金融": "finance",
+        "经济": "finance"
+    }
+
+    # 转换为小写并去除空格
+    normalized_category = category_text.lower().strip()
+
+    # 返回映射结果，如果没有匹配项则返回原始输入
+    return category_map.get(normalized_category, category_text)
+
+
+@register_function('get_news', GET_NEWS_FUNCTION_DESC, ToolType.SYSTEM_CTL)
+def get_news(conn, category: str = None, detail: bool = False, lang: str = "zh_CN"):
+    """获取新闻并随机选择一条进行播报，或获取上一条新闻的详细内容"""
+    try:
+        # 如果detail为True，获取上一条新闻的详细内容
+        if detail:
+            if not hasattr(conn, 'last_news_link') or not conn.last_news_link or 'link' not in conn.last_news_link:
+                return ActionResponse(Action.REQLLM, "抱歉，没有找到最近查询的新闻，请先获取一条新闻。", None)
+
+            link = conn.last_news_link.get('link')
+            title = conn.last_news_link.get('title', '未知标题')
+
+            if link == '#':
+                return ActionResponse(Action.REQLLM, "抱歉，该新闻没有可用的链接获取详细内容。", None)
+
+            logger.bind(tag=TAG).debug(f"获取新闻详情: {title}, URL={link}")
+
+            # 获取新闻详情
+            detail_content = fetch_news_detail(link)
+
+            if not detail_content or detail_content == "无法获取详细内容":
+                return ActionResponse(Action.REQLLM,
+                                      f"抱歉，无法获取《{title}》的详细内容，可能是链接已失效或网站结构发生变化。", None)
+
+            # 构建详情报告
+            detail_report = (
+                f"根据下列数据，用{lang}回应用户的新闻详情查询请求：\n\n"
+                f"新闻标题: {title}\n"
+                f"详细内容: {detail_content}\n\n"
+                f"(请对上述新闻内容进行总结，提取关键信息，以自然、流畅的方式向用户播报，"
+                f"不要提及这是总结，就像是在讲述一个完整的新闻故事)"
+            )
+
+            return ActionResponse(Action.REQLLM, detail_report, None)
+
+        # 否则，获取新闻列表并随机选择一条
+        # 从配置中获取RSS URL
+        rss_config = conn.config["plugins"]["get_news"]
+        default_rss_url = rss_config.get("default_rss_url", "https://www.chinanews.com.cn/rss/society.xml")
+
+        # 将用户输入的类别映射到配置中的类别键
+        mapped_category = map_category(category)
+
+        # 如果提供了类别，尝试从配置中获取对应的URL
+        rss_url = default_rss_url
+        if mapped_category and mapped_category in rss_config.get("category_urls", {}):
+            rss_url = rss_config["category_urls"][mapped_category]
+
+        logger.bind(tag=TAG).info(f"获取新闻: 原始类别={category}, 映射类别={mapped_category}, URL={rss_url}")
+
+        # 获取新闻列表
+        news_items = fetch_news_from_rss(rss_url)
+
+        if not news_items:
+            return ActionResponse(Action.REQLLM, "抱歉，未能获取到新闻信息，请稍后再试。", None)
+
+        # 随机选择一条新闻
+        selected_news = random.choice(news_items)
+
+        # 保存当前新闻链接到连接对象，以便后续查询详情
+        if not hasattr(conn, 'last_news_link'):
+            conn.last_news_link = {}
+        conn.last_news_link = {
+            'link': selected_news.get('link', '#'),
+            'title': selected_news.get('title', '未知标题')
+        }
+
+        # 构建新闻报告
+        news_report = (
+            f"根据下列数据，用{lang}回应用户的新闻查询请求：\n\n"
+            f"新闻标题: {selected_news['title']}\n"
+            f"发布时间: {selected_news['pubDate']}\n"
+            f"新闻内容: {selected_news['description']}\n"
+            f"(请以自然、流畅的方式向用户播报这条新闻，可以适当总结内容，"
+            f"直接读出新闻即可，不需要额外多余的内容。"
+            f"如果用户询问更多详情，告知用户可以说'请详细介绍这条新闻'获取更多内容)"
+        )
+
+        return ActionResponse(Action.REQLLM, news_report, None)
+
+    except Exception as e:
+        logger.bind(tag=TAG).error(f"获取新闻出错: {e}")
+        return ActionResponse(Action.REQLLM, "抱歉，获取新闻时发生错误，请稍后再试。", None)

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

@@ -1,20 +1,19 @@
 from datetime import datetime
+import cnlunar
 from plugins_func.register import register_function, ToolType, ActionResponse, Action
 
 get_time_function_desc = {
     "type": "function",
     "function": {
         "name": "get_time",
-        "description": "获取当前时间、日期、星期几",
+        "description": "获取今天日期或者当前时间信息",
         'parameters': {'type': 'object', 'properties': {}, 'required': []}
     }
 }
-
-
 @register_function('get_time', get_time_function_desc, ToolType.WAIT)
 def get_time():
     """
-    获取当前时间、日期、星期几
+    获取当前的日期时间信息
     """
     now = datetime.now()
     current_time = now.strftime("%H:%M:%S")
@@ -22,4 +21,69 @@ def get_time():
     current_weekday = now.strftime("%A")
     response_text = f"当前日期: {current_date}，当前时间: {current_time}，星期: {current_weekday}"
 
+    return ActionResponse(Action.REQLLM, response_text, None)
+
+
+get_lunar_function_desc = {
+    "type": "function",
+    "function": {
+        "name": "get_lunar",
+        "description": (
+          "用于获取今天的阴历/农历和黄历信息。"
+          "用户可以指定查询内容，如：阴历日期、天干地支、节气、生肖、星座、八字、宜忌等。"
+          "如果没有指定查询内容，则默认查询干支年和农历日期。"
+        ),
+        "parameters": {
+            "type": "object", 
+            "properties": {
+                "query": {
+                    "type": "string",
+                    "description": "要查询的内容，例如阴历日期、天干地支、节日、节气、生肖、星座、八字、宜忌等"
+                }
+            }, 
+            "required": []
+        }
+    }
+}
+@register_function('get_lunar', get_lunar_function_desc, ToolType.WAIT)
+def get_lunar(query=None):
+    """
+    用于获取当前的阴历/农历，和天干地支、节气、生肖、星座、八字、宜忌等黄历信息
+    """
+    now = datetime.now()
+    current_time = now.strftime("%H:%M:%S")
+    current_date = now.strftime("%Y-%m-%d")
+    current_weekday = now.strftime("%A")
+
+    # 如果 query 为 None，则使用默认文本
+    if query is None:
+        query = "默认查询干支年和农历日期"
+    response_text = f"根据以下信息回应用户的查询请求，并提供与{query}相关的信息：\n"
+
+    lunar = cnlunar.Lunar(now, godType='8char')
+    response_text += (
+        f"当前公历日期: {current_date}，当前时间: {current_time}，星期: {current_weekday}\n"
+        "农历信息：\n"
+        "%s年%s%s\n" % (lunar.lunarYearCn, lunar.lunarMonthCn[:-1], lunar.lunarDayCn) +
+        "干支: %s年 %s月 %s日\n" % (lunar.year8Char, lunar.month8Char, lunar.day8Char) +
+        "生肖: 属%s\n" % (lunar.chineseYearZodiac) +
+        "八字: %s\n" % (' '.join([lunar.year8Char, lunar.month8Char, lunar.day8Char, lunar.twohour8Char])) +
+        "今日节日: %s\n" % (",".join(filter(None, (lunar.get_legalHolidays(), lunar.get_otherHolidays(), lunar.get_otherLunarHolidays())))) +
+        "今日节气: %s\n" % (lunar.todaySolarTerms) +
+        "下一节气: %s %s年%s月%s日\n" % (lunar.nextSolarTerm, lunar.nextSolarTermYear, lunar.nextSolarTermDate[0], lunar.nextSolarTermDate[1]) +
+        "今年节气表: %s\n" % (', '.join([f"{term}({date[0]}月{date[1]}日)" for term, date in lunar.thisYearSolarTermsDic.items()])) +
+        "生肖冲煞: %s\n" % (lunar.chineseZodiacClash) +
+        "星座: %s\n" % (lunar.starZodiac) +
+        "纳音: %s\n" % lunar.get_nayin() +
+        "彭祖百忌: %s\n" % (lunar.get_pengTaboo(delimit=", ")) +
+        "值日: %s执位\n" % lunar.get_today12DayOfficer()[0] +
+        "值神: %s(%s)\n" % (lunar.get_today12DayOfficer()[1], lunar.get_today12DayOfficer()[2]) +
+        "廿八宿: %s\n" % lunar.get_the28Stars() +
+        "吉神方位: %s\n" % ' '.join(lunar.get_luckyGodsDirection()) +
+        "今日胎神: %s\n" % lunar.get_fetalGod() +
+        "宜: %s\n" % '、'.join(lunar.goodThing[:10]) +
+        "忌: %s\n" % '、'.join(lunar.badThing[:10]) +
+        "(默认返回干支年和农历日期；仅在要求查询宜忌信息时才返回本日宜忌)"
+    )
+
     return ActionResponse(Action.REQLLM, response_text, None)

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

@@ -39,6 +39,23 @@ HEADERS = {
     )
 }
 
+# 天气代码 https://dev.qweather.com/docs/resource/icons/#weather-icons
+WEATHER_CODE_MAP = {
+    "100": "晴", "101": "多云", "102": "少云", "103": "晴间多云", "104": "阴",
+    "150": "晴", "151": "多云", "152": "少云", "153": "晴间多云",
+    "300": "阵雨", "301": "强阵雨", "302": "雷阵雨", "303": "强雷阵雨", "304": "雷阵雨伴有冰雹",
+    "305": "小雨", "306": "中雨", "307": "大雨", "308": "极端降雨", "309": "毛毛雨/细雨",
+    "310": "暴雨", "311": "大暴雨", "312": "特大暴雨", "313": "冻雨", "314": "小到中雨",
+    "315": "中到大雨", "316": "大到暴雨", "317": "暴雨到大暴雨", "318": "大暴雨到特大暴雨",
+    "350": "阵雨", "351": "强阵雨", "399": "雨",
+    "400": "小雪", "401": "中雪", "402": "大雪", "403": "暴雪", "404": "雨夹雪",
+    "405": "雨雪天气", "406": "阵雨夹雪", "407": "阵雪", "408": "小到中雪", "409": "中到大雪", "410": "大到暴雪",
+    "456": "阵雨夹雪", "457": "阵雪", "499": "雪",
+    "500": "薄雾", "501": "雾", "502": "霾", "503": "扬沙", "504": "浮尘",
+    "507": "沙尘暴", "508": "强沙尘暴",
+    "509": "浓雾", "510": "强浓雾", "511": "中度霾", "512": "重度霾", "513": "严重霾", "514": "大雾", "515": "特强浓雾",
+    "900": "热", "901": "冷", "999": "未知"
+}
 
 def fetch_city_info(location, api_key):
     url = f"https://geoapi.qweather.com/v2/city/lookup?key={api_key}&location={location}&lang=zh"
@@ -67,9 +84,11 @@ def parse_weather_info(soup):
     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)
+        weather_code = row.select_one(".date-bg .icon")["src"].split("/")[-1].split(".")[0]
+        weather = WEATHER_CODE_MAP.get(weather_code, "未知")
         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))
+        temps_list.append((date, weather, high_temp, low_temp))
 
     return city_name, current_abstract, current_basic, temps_list
 
@@ -91,14 +110,14 @@ def get_weather(conn, location: str = None, lang: str = "zh_CN"):
 
     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):
+    for i, (date, weather, high, low) in enumerate(temps_list):
         if high and low:
-            weather_report += f"{date}: {low}到{high}\n"
+            weather_report += f"{date}: {low}到{high}, {weather}\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。"
+        f"(确保只报告指定单日的天气情况，除非未来会出现异常天气；或者用户明确要求想要了解多日天气，如果未指定，默认报告今天的天气。"
         "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)
+    return ActionResponse(Action.REQLLM, weather_report, None)

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

@@ -0,0 +1,111 @@
+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()
+
+
+async def _get_device_status(conn, device_name, device_type, property_name):
+    """获取设备状态"""
+    status = await get_iot_status(conn, device_type, property_name)
+    if status is None:
+        raise Exception(f"你的设备不支持{device_name}控制")
+    return status
+
+
+async def _set_device_property(conn, device_name, device_type, method_name, property_name, new_value=None, action=None, step=10):
+    """设置设备属性"""
+    current_value = await _get_device_status(conn, device_name, device_type, property_name)
+
+    if action == 'raise':
+        current_value += step
+    elif action == 'lower':
+        current_value -= step
+    elif action == 'set':
+        if new_value is None:
+            raise Exception(f"缺少{property_name}参数")
+        current_value = new_value
+
+    # 限制属性范围在0到100之间
+    current_value = max(0, min(100, current_value))
+
+    await send_iot_conn(conn, device_type, method_name, {property_name: current_value})
+    return current_value
+
+
+def _handle_device_action(conn, func, success_message, error_message, *args, **kwargs):
+    """处理设备操作的通用函数"""
+    future = asyncio.run_coroutine_threadsafe(
+        func(conn, *args, **kwargs), conn.loop)
+    try:
+        result = future.result()
+        logger.bind(tag=TAG).info(f"{success_message}: {result}")
+        response = f"{success_message}{result}"
+        return ActionResponse(action=Action.RESPONSE, result=result, response=response)
+    except Exception as e:
+        logger.bind(tag=TAG).error(f"{error_message}: {e}")
+        response = f"{error_message}: {e}"
+        return ActionResponse(action=Action.RESPONSE, result=None, response=response)
+
+
+# 设备控制
+handle_device_function_desc = {
+    "type": "function",
+    "function": {
+        "name": "handle_device",
+        "description": (
+            "用户想要获取或者设置设备的音量/亮度大小，或者用户觉得声音/亮度过高或过低，或者用户想提高或降低音量/亮度。"
+            "比如用户说现在亮度多少，参数为：device_type:Screen,action:get。"
+            "比如用户说设置音量为50，参数为：device_type:Speaker,action:set,value:50。"
+            "比如用户说亮度太高了，参数为：device_type:Screen,action:lower。"
+            "比如用户说调大音量，参数为：device_type:Speaker,action:raise。"
+        ),
+        "parameters": {
+            "type": "object",
+            "properties": {
+                "device_type": {
+                    "type": "string",
+                    "description": "设备类型，可选值：Speaker(音量),Screen(亮度)"
+                },
+                "action": {
+                    "type": "string",
+                    "description": "动作名称，可选值：get(获取),set(设置),raise(提高),lower(降低)"
+                },
+                "value": {
+                    "type": "integer",
+                    "description": "值大小，可选值：0-100之间的整数"
+                }
+            },
+            "required": ["device_type", "action"]
+        }
+    }
+}
+
+
+@register_function('handle_device', handle_device_function_desc, ToolType.IOT_CTL)
+def handle_device(conn, device_type: str, action: str, value: int = None):
+    if device_type == "Speaker":
+        method_name, property_name, device_name = "SetVolume", "volume", "音量"
+    elif device_type == "Screen":
+        method_name, property_name, device_name = "SetBrightness", "brightness", "亮度"
+    else:
+        raise Exception(f"未识别的设备类型: {device_type}")
+
+    if action not in ["get", "set", "raise", "lower"]:
+        raise Exception(f"未识别的动作名称: {action}")
+
+    if action == "get":
+        # get
+        return _handle_device_action(
+            conn, _get_device_status, f"当前{device_name}", f"获取{device_name}失败",
+            device_name=device_name, device_type=device_type, property_name=property_name,
+        )
+    else:
+        # set, raise, lower
+        return _handle_device_action(
+            conn, _set_device_property, f"{device_name}已调整到", f"{device_name}调整失败",
+            device_name=device_name, device_type=device_type, method_name=method_name,
+            property_name=property_name, new_value=value, action=action
+        )

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

@@ -0,0 +1,56 @@
+from plugins_func.register import register_function, ToolType, ActionResponse, Action
+from plugins_func.functions.hass_init import initialize_hass_handler
+from config.logger import setup_logging
+import asyncio
+import requests
+
+TAG = __name__
+logger = setup_logging()
+
+hass_get_state_function_desc = {
+    "type": "function",
+    "function": {
+        "name": "hass_get_state",
+        "description": "获取homeassistant里设备的状态,包括灯光亮度,媒体播放器的音量,设备的暂停、继续操作",
+        "parameters": {
+            "type": "object",
+            "properties": {
+                "entity_id": {
+                    "type": "string",
+                    "description": "需要操作的设备id,homeassistant里的entity_id"
+                }
+            },
+            "required": ["entity_id"]
+        }
+    }
+}
+
+
+@register_function("hass_get_state", hass_get_state_function_desc, ToolType.SYSTEM_CTL)
+def hass_get_state(conn, entity_id=''):
+    try:
+
+        future = asyncio.run_coroutine_threadsafe(
+            handle_hass_get_state(conn, entity_id),
+            conn.loop
+        )
+        ha_response = future.result()
+        return ActionResponse(action=Action.REQLLM, result="执行成功", response=ha_response)
+    except Exception as e:
+        logger.bind(tag=TAG).error(f"处理设置属性意图错误: {e}")
+
+
+async def handle_hass_get_state(conn, entity_id):
+    HASS_CACHE = initialize_hass_handler(conn)
+    api_key = HASS_CACHE['api_key']
+    base_url = HASS_CACHE['base_url']
+    url = f"{base_url}/api/states/{entity_id}"
+    headers = {
+        "Authorization": f"Bearer {api_key}",
+        "Content-Type": "application/json"
+    }
+    response = requests.get(url, headers=headers)
+    if response.status_code == 200:
+        return response.json()['state']
+    else:
+        return f"切换失败，错误码: {response.status_code}"

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

@@ -0,0 +1,64 @@
+from plugins_func.register import register_function, ToolType, ActionResponse, Action
+from plugins_func.functions.hass_init import initialize_hass_handler
+from config.logger import setup_logging
+import asyncio
+import requests
+
+TAG = __name__
+logger = setup_logging()
+
+hass_play_music_function_desc = {
+    "type": "function",
+    "function": {
+        "name": "hass_play_music",
+        "description": "用户想听音乐、有声书的时候使用，在房间的媒体播放器（media_player）里播放对应音频",
+        "parameters": {
+            "type": "object",
+            "properties": {
+                "media_content_id": {
+                    "type": "string",
+                    "description": "可以是音乐或有声书的专辑名称、歌曲名、演唱者,如果未指定就填random"
+                },
+                "entity_id": {
+                    "type": "string",
+                    "description": "需要操作的音箱的设备id,homeassistant里的entity_id,media_player开头"
+                }
+            },
+            "required": ["media_content_id", "entity_id"]
+        }
+    }
+}
+
+
+@register_function('hass_play_music', hass_play_music_function_desc, ToolType.SYSTEM_CTL)
+def hass_play_music(conn, entity_id='', media_content_id='random'):
+    try:
+        # 执行音乐播放命令
+        future = asyncio.run_coroutine_threadsafe(
+            handle_hass_play_music(conn, entity_id, media_content_id),
+            conn.loop
+        )
+        ha_response = future.result()
+        return ActionResponse(action=Action.RESPONSE, result="退出意图已处理", response=ha_response)
+    except Exception as e:
+        logger.bind(tag=TAG).error(f"处理音乐意图错误: {e}")
+
+
+async def handle_hass_play_music(conn, entity_id, media_content_id):
+    HASS_CACHE = initialize_hass_handler(conn)
+    api_key = HASS_CACHE['api_key']
+    base_url = HASS_CACHE['base_url']
+    url = f"{base_url}/api/services/music_assistant/play_media"
+    headers = {
+        "Authorization": f"Bearer {api_key}",
+        "Content-Type": "application/json"
+    }
+    data = {
+        "entity_id": entity_id,
+        "media_id": media_content_id
+    }
+    response = requests.post(url, headers=headers, json=data)
+    if response.status_code == 200:
+        return f"正在播放{media_content_id}的音乐"
+    else:
+        return f"音乐播放失败，错误码: {response.status_code}"

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

@@ -0,0 +1,159 @@
+from plugins_func.register import register_function, ToolType, ActionResponse, Action
+from plugins_func.functions.hass_init import initialize_hass_handler
+from config.logger import setup_logging
+import asyncio
+import requests
+
+TAG = __name__
+logger = setup_logging()
+
+hass_set_state_function_desc = {
+    "type": "function",
+    "function": {
+        "name": "hass_set_state",
+        "description": "设置homeassistant里设备的状态,包括开、关,调整灯光亮度,调整播放器的音量,设备的暂停、继续、静音操作",
+        "parameters": {
+            "type": "object",
+            "properties": {
+                "state": {
+                    "type": "object",
+                    "properties": {
+                        "type": {
+                            "type": "string",
+                            "description": "需要操作的动作,打开设备:turn_on,关闭设备:turn_off,增加亮度:brightness_up,降低亮度:brightness_down,设置亮度:brightness_value,增加>音量:,volume_up降低音量:volume_down,设置音量:volume_set,设备暂停:pause,设备继续:continue,静音/取消静音:volume_mute"
+                        },
+                        "input": {
+                            "type": "integer",
+                            "description": "只有在设置音量,设置亮度时候才需要,有效值为1-100,对应音量和亮度的1%-100%"
+                        },
+                        "is_muted": {
+                            "type": "string",
+                            "description": "只有在设置静音操作时才需要,设置静音的时候该值为true,取消静音时该值为false"
+                        }
+                    },
+                    "required": ["type"]
+                },
+                "entity_id": {
+                    "type": "string",
+                    "description": "需要操作的设备id,homeassistant里的entity_id"
+                }
+            },
+            "required": ["state", "entity_id"]
+        }
+    }
+}
+
+
+@register_function('hass_set_state', hass_set_state_function_desc, ToolType.SYSTEM_CTL)
+def hass_set_state(conn, entity_id='', state={}):
+    try:
+        future = asyncio.run_coroutine_threadsafe(
+            handle_hass_set_state(conn, entity_id, state),
+            conn.loop
+        )
+        ha_response = future.result()
+        return ActionResponse(action=Action.REQLLM, result="执行成功", response=ha_response)
+    except Exception as e:
+        logger.bind(tag=TAG).error(f"处理设置属性意图错误: {e}")
+
+
+async def handle_hass_set_state(conn, entity_id, state):
+    HASS_CACHE = initialize_hass_handler(conn)
+    api_key = HASS_CACHE['api_key']
+    base_url = HASS_CACHE['base_url']
+    '''
+    state = { "type":"brightness_up","input":"80","is_muted":"true"}
+    '''
+    domains = entity_id.split(".")
+    if len(domains) > 1:
+        domain = domains[0]
+    else:
+        return "执行失败，错误的设备id"
+    action = ''
+    arg = ''
+    value = ''
+    if state['type'] == 'turn_on':
+        description = "设备已打开"
+        if domain == "cover":
+            action = "open_cover"
+        elif domain == "vacuum":
+            action = "start"
+        else:
+            action = "turn_on"
+    elif state['type'] == 'turn_off':
+        description = "设备已关闭"
+        if domain == 'cover':
+            action = "close_cover"
+        elif domain == 'vacuum':
+            action = "stop"
+        else:
+            action = "turn_off"
+    elif state['type'] == 'brightness_up':
+        description = "灯光已调亮"
+        action = 'turn_on'
+        arg = 'brightness_step_pct'
+        value = 10
+    elif state['type'] == 'brightness_down':
+        description = "灯光已调暗"
+        action = 'turn_on'
+        arg = 'brightness_step_pct'
+        value = -10
+    elif state['type'] == 'brightness_value':
+        description = f"亮度已调整到{state['input']}"
+        action = 'turn_on'
+        arg = 'brightness_pct'
+        value = state['input']
+    elif state['type'] == 'volume_up':
+        description = "音量已调大"
+        action = state['type']
+    elif state['type'] == 'volume_down':
+        description = "音量已调小"
+        action = state['type']
+    elif state['type'] == 'volume_set':
+        description = f"音量已调整到{state['input']}"
+        action = state['type']
+        arg = 'volume_level'
+        value = state['input']
+    elif state['type'] == 'volume_mute':
+        description = f"设备已静音"
+        action = state['type']
+        arg = 'is_volume_muted'
+        value = state['is_muted']
+    elif state['type'] == 'pause':
+        description = f"设备已暂停"
+        action = state['type']
+        if domain == 'media_player':
+            action = 'media_pause'
+        if domain == 'cover':
+            action = 'stop_cover'
+        if domain == 'vacuum':
+            action = 'pause'
+    elif state['type'] == 'continue':
+        description = f"设备已继续"
+        if domain == 'media_player':
+            action = 'media_play'
+        if domain == 'vacuum':
+            action = 'start'
+    else:
+        return f"{domain} {state.type}功能尚未支持"
+
+    if arg == '':
+        data = {
+            "entity_id": entity_id,
+        }
+    else:
+        data = {
+            "entity_id": entity_id,
+            arg: value
+        }
+    url = f"{base_url}/api/services/{domain}/{action}"
+    headers = {
+        "Authorization": f"Bearer {api_key}",
+        "Content-Type": "application/json"
+    }
+    response = requests.post(url, headers=headers, json=data)
+    logger.bind(tag=TAG).info(f"设置状态:url:{url},return_code:{response.status_code}")
+    if response.status_code == 200:
+        return description
+    else:
+        return f"设置失败，错误码: {response.status_code}"

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

@@ -217,13 +217,14 @@ async def play_local_music(conn, specific_file=None):
         )
         conn.tts.tts_audio_queue.put(
             TTSMessageDTO(
-                u_id=conn.u_id, msg_type=MsgType.STOP_TTS_RESPONSE, content=[],
+                u_id=conn.u_id,
-                tts_finish_text='',
+                msg_type=MsgType.STOP_TTS_RESPONSE,
-                sentence_type=None
+                content=[],
+                tts_finish_text="",
+                sentence_type=None,
             )
         )
 
-
     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()}")

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

@@ -1,64 +0,0 @@
-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

main/xiaozhi-server/plugins_func/loadplugins.py

@@ -22,6 +22,4 @@ def auto_import_modules(package_name):
         # 导入模块
         full_module_name = f"{package_name}.{module_name}"
         importlib.import_module(full_module_name)
-        #logger.bind(tag=TAG).info(f"模块 '{full_module_name}' 已加载")
+        #logger.bind(tag=TAG).info(f"模块 '{full_module_name}' 已加载")
-
-auto_import_modules('plugins_func.functions')

main/xiaozhi-server/plugins_func/register.py

@@ -12,6 +12,7 @@ class ToolType(Enum):
     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")
+    MCP_CLIENT = (6, "MCP客户端")
 
     def __init__(self, code, message):
         self.code = code

main/xiaozhi-server/requirements.txt

@@ -22,4 +22,5 @@ mem0ai==0.1.62
 bs4==0.0.2
 modelscope==1.23.2
 sherpa_onnx==1.11.0
-mutagen==1.47.0
+mcp==1.4.1
+cnlunar==0.2.0