update:homeassistant插件化改造

update:homeassistant插件化改造

README.md

@@ -51,7 +51,7 @@
         </a>
     </td>
     <td>
-        <a href="https://www.bilibili.com/video/av114036381327149" target="_blank">
+        <a href="https://www.bilibili.com/video/BV1pNXWYGEx1" target="_blank">
          <picture>
            <img alt="控制家电开关" src="docs/images/demo5.png" />
          </picture>
@@ -95,6 +95,11 @@
         </a>
     </td>
     <td>
+        <a href="https://www.bilibili.com/video/BV17LXWYvENb" target="_blank">
+         <picture>
+           <img alt="播报新闻" src="docs/images/demo0.png" />
+         </picture>
+        </a>
     </td>
   </tr>
 </table>
@@ -185,7 +190,6 @@ server:
 | LLM |     FastgptLLM     |     fastgpt 接口调用      | 免费/消耗 token |                                              本地化部署，注意配置提示词需在 Fastgpt 控制台设置                                              |
 | LLM |     GeminiLLM      |      gemini 接口调用      |     免费      |                                      [点击申请密钥](https://aistudio.google.com/apikey)                                       |
 | LLM |      CozeLLM       |       coze 接口调用       |  消耗 token   |                                                需提供 bot_id、user_id 及个人令牌                                                 |
-| LLM |   Home Assistant   | homeassistant语音助手接口调用 |     免费      |                                                   需提供home assistant令牌                                                   |
 
 实际上，任何支持 openai 接口调用的 LLM 均可接入使用。
 
@@ -370,7 +374,41 @@ VAD:
 
 ### 6、我想通过小智控制电灯、空调、远程开关机等操作 💡
 
-建议：在配置文件中将 `LLM` 设置为 `HomeAssistant`，通过 调用`HomeAssistant`接口实现相关控制。
+本项目，支持以工具调用的方式控制HomeAssistant设备
+
+1、首先选择一款支持function call支持的LLM，例如`ChatGLMLLM`。
+
+2、在配置文件中，将 `selected_module.Intent` 设置为 `function_call`。
+
+3、登录`HomeAssistant`，点击`左下角个人`，切换`安全`导航栏，划到底部`长期访问令牌`生成api_key。
+
+在配置文件中，配置好你的home assistant的`devices`（被控制的设备）和`api_key`和`base_url`等信息。例如：
+
+``` yaml 
+plugins
+  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://你的homeassistant地址:8123
+    api_key: 你的home assistant api访问令牌
+```
+
+最后，允许function_call 插件在配置文件中启用`hass_get_state`(必须)、`hass_set_state`(必须)、`hass_play_music`(不想用ha听音乐可以不启动)，例如：
+
+``` yaml 
+Intent:
+  ...
+  function_call:
+    type: nointent
+    functions:
+      - change_role
+      - get_weather
+      - get_news
+      - hass_get_state
+      - hass_set_state
+      - hass_play_music
+```
 
 ### 7、更多问题，可联系我们反馈 💬
 

main/xiaozhi-server/config.yaml

@@ -49,7 +49,7 @@ 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。
   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。
   现在我正在和你进行语音聊天，我们开始吧。
-  如果用户希望结束对话，请在最后说“拜拜”或“再见”。
+
 # 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
 
@@ -99,6 +99,12 @@ Intent:
       - 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:
@@ -115,6 +121,13 @@ plugins:
       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
+
 
 Memory:
   mem0ai:
@@ -241,12 +254,6 @@ 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
@@ -500,20 +507,3 @@ module_test:
     - "你好，请介绍一下你自己"
     - "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
-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
-
-# 以下配置在小于等于0.0.9版本中的docker容器中可用
-# 0.0.9以后的新版本源码部署已经无法奏效
-manager:
-  enabled: false
-  ip: 0.0.0.0
-  port: 8002
-use_private_config: false

main/xiaozhi-server/core/connection.py

@@ -196,7 +196,7 @@ class ConnectionHandler:
         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)
+        self.func_handler = FunctionHandler(self)
 
     def change_system_prompt(self, prompt):
         self.prompt = prompt

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

@@ -1,16 +1,16 @@
-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):
-        self.config = config
+    def __init__(self, conn):
+        self.conn = conn
+        self.config = conn.config
         self.function_registry = FunctionRegistry()
         self.register_nessary_functions()
         self.register_config_functions()
@@ -26,9 +26,10 @@ 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()
 
@@ -47,7 +48,6 @@ 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")
@@ -57,6 +57,9 @@ class FunctionHandler:
         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)
 
@@ -81,4 +84,4 @@ class FunctionHandler:
         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/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/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/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_init.py

@@ -0,0 +1,35 @@
+from config.logger import setup_logging
+from core.utils.util import check_model_key
+
+TAG = __name__
+logger = setup_logging()
+
+HASS_CACHE = {}
+
+
+def append_devices_to_prompt(conn):
+    if conn.use_function_call_mode:
+        funcs = conn.config["Intent"]["function_call"].get("functions", [])
+        if "hass_get_state" in funcs or "hass_get_state" in funcs:
+            prompt = "下面是我家智能设备，可以通过homeassistant控制\n"
+            devices = conn.config["plugins"]["home_assistant"].get("devices", [])
+            if len(devices) == 0:
+                return
+            for device in devices:
+                prompt += device + "\n"
+            conn.prompt += prompt
+            # 更新提示词
+            conn.dialogue.update_system_message(conn.prompt)
+
+
+def initialize_hass_handler(conn):
+    global HASS_CACHE
+    if HASS_CACHE == {}:
+        if conn.use_function_call_mode:
+            funcs = conn.config["Intent"]["function_call"].get("functions", [])
+            if "hass_get_state" in funcs or "hass_get_state" in funcs:
+                HASS_CACHE['base_url'] = conn.config["plugins"]["home_assistant"].get("base_url")
+                HASS_CACHE['api_key'] = conn.config["plugins"]["home_assistant"].get("api_key")
+
+                check_model_key("home_assistant", HASS_CACHE['api_key'])
+    return HASS_CACHE

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": "int",
+                            "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}"