update:优化唤醒词答复

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

@@ -12,19 +12,21 @@ from core.handle.mcpHandle import (
     send_mcp_initialize_message,
     send_mcp_tools_list_request,
 )
-
+from core.utils.wakeup_word import WakeupWordsConfig
 
 TAG = __name__
 
 WAKEUP_CONFIG = {
-    "dir": "config/assets/",
-    "file_name": "wakeup_words",
-    "create_time": time.time(),
-    "refresh_time": 10,
+    "refresh_time": 5,
     "words": ["你好小智", "你好啊小智", "小智你好", "小智"],
-    "text": "",
 }
 
+# 创建全局的唤醒词配置管理器
+wakeup_words_config = WakeupWordsConfig()
+
+# 用于防止并发调用wakeupWordsResponse的锁
+_wakeup_response_lock = asyncio.Lock()
+
 
 async def handleHelloMessage(conn, msg_json):
     """处理hello消息"""
@@ -55,96 +57,78 @@ async def checkWakeupWords(conn, text):
     enable_wakeup_words_response_cache = conn.config[
         "enable_wakeup_words_response_cache"
     ]
-    """是否用的是非流式tts"""
-    if conn.tts and conn.tts.interface_type != InterfaceType.NON_STREAM:
+
+    if not enable_wakeup_words_response_cache or not conn.tts:
         return False
 
-    """是否开启唤醒词加速"""
-    if not enable_wakeup_words_response_cache:
+    if conn.tts.interface_type != InterfaceType.NON_STREAM:
         return False
-    """检查是否是唤醒词"""
+
     _, filtered_text = remove_punctuation_and_length(text)
-    if filtered_text in conn.config.get("wakeup_words"):
-        await send_stt_message(conn, text)
+    if filtered_text not in conn.config.get("wakeup_words"):
+        return False
 
-        file = getWakeupWordFile(WAKEUP_CONFIG["file_name"])
-        if file is None:
+    await send_stt_message(conn, text)
+
+    # 获取当前音色
+    voice = getattr(conn.tts, "voice", "default")
+
+    # 获取唤醒词回复配置
+    response = wakeup_words_config.get_wakeup_response(voice)
+
+    # 播放唤醒词回复
+    conn.client_abort = False
+    conn.tts.tts_one_sentence(
+        conn,
+        ContentType.FILE,
+        content_file=response["file_path"],
+        content_detail=response["text"],
+    )
+
+    # 检查是否需要更新唤醒词回复
+    if time.time() - response["time"] > WAKEUP_CONFIG["refresh_time"]:
+        if not _wakeup_response_lock.locked():
             asyncio.create_task(wakeupWordsResponse(conn))
-            return False
-        text_hello = WAKEUP_CONFIG["text"]
-        if not text_hello:
-            text_hello = text
-        if conn.tts is None:
-            return False
-        conn.tts.tts_one_sentence(
-            conn, ContentType.FILE, content_file=file, content_detail=text_hello
-        )
-        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
+    return True
 
 
 async def wakeupWordsResponse(conn):
-    wait_max_time = 5
-    while conn.llm is None or not conn.llm.response_no_stream:
-        await asyncio.sleep(1)
-        wait_max_time -= 1
-        if wait_max_time <= 0:
-            conn.logger.bind(tag=TAG).error("连接对象没有llm")
-            return
-
-    """唤醒词响应"""
-    wakeup_word = random.choice(WAKEUP_CONFIG["words"])
-    question = (
-        "此刻用户正在和你说```"
-        + wakeup_word
-        + "```。\n请你根据以上用户的内容，进行简短回复，文字内容控制在15个字以内。\n"
-        + "请勿对这条内容本身进行任何解释和回应，仅返回对用户的内容的回复。"
-    )
-    result = conn.llm.response_no_stream(conn.config["prompt"], question)
-    if result is None or result == "":
+    if not conn.tts or not conn.llm or not conn.llm.response_no_stream:
         return
 
-    tts_result = await asyncio.to_thread(conn.tts.to_tts, result)
-    if conn.tts.delete_audio_file:
-        # 返回的是opus数据流
+    try:
+        # 尝试获取锁，如果获取不到就返回
+        if not await _wakeup_response_lock.acquire():
+            return
+
+        # 生成唤醒词回复
+        wakeup_word = random.choice(WAKEUP_CONFIG["words"])
+        question = (
+            "此刻用户正在和你说```"
+            + wakeup_word
+            + "```。\n请你根据以上用户的内容进行简短回复。要像一个人正常人一样说话，不要像机器人一样说话。\n"
+            + "请勿对这条内容本身进行任何解释和回应，请勿返回表情符号，仅返回对用户的内容的回复。"
+        )
+
+        result = conn.llm.response_no_stream(conn.config["prompt"], question)
+        if not result or len(result) == 0:
+            return
+
+        # 生成TTS音频
+        tts_result = await asyncio.to_thread(conn.tts.to_tts, result)
         if not tts_result:
             return
+
+        # 获取当前音色
+        voice = getattr(conn.tts, "voice", "default")
+
         wav_bytes = opus_datas_to_wav_bytes(tts_result, sample_rate=16000)
-        file_path = os.path.join(
-            WAKEUP_CONFIG["dir"], "my_" + WAKEUP_CONFIG["file_name"] + ".wav"
-        )
+        file_path = wakeup_words_config.generate_file_path(voice)
         with open(file_path, "wb") as f:
             f.write(wav_bytes)
-
-    else:
-        # 返回为文件
-        if tts_result is not None and os.path.exists(tts_result):
-            file_type = os.path.splitext(tts_result)[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目录下"""
-            shutil.move(
-                tts_result,
-                WAKEUP_CONFIG["dir"] + "my_" + WAKEUP_CONFIG["file_name"] + '.' + file_type,
-            )
-    WAKEUP_CONFIG["create_time"] = time.time()
-    WAKEUP_CONFIG["text"] = result
+        # 更新配置
+        wakeup_words_config.update_wakeup_response(voice, file_path, result)
+    finally:
+        # 确保在任何情况下都释放锁
+        if _wakeup_response_lock.locked():
+            _wakeup_response_lock.release()

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

@@ -145,10 +145,9 @@ class TTSProvider(TTSProviderBase):
         self.cluster = config.get("cluster")
         self.resource_id = config.get("resource_id")
         if config.get("private_voice"):
-            self.speaker = config.get("private_voice")
+            self.voice = config.get("private_voice")
         else:
-            self.speaker = config.get("speaker")
-        self.voice = config.get("voice")
+            self.voice = config.get("speaker")
         self.ws_url = config.get("ws_url")
         self.authorization = config.get("authorization")
         self.header = {"Authorization": f"{self.authorization}{self.access_token}"}
@@ -272,7 +271,7 @@ class TTSProvider(TTSProviderBase):
                 logger.bind(tag=TAG).error(f"WebSocket连接不存在，终止发送文本")
                 return
             # 发送文本
-            await self.send_text(self.speaker, text, self.conn.sentence_id)
+            await self.send_text(self.voice, text, self.conn.sentence_id)
             return
         except Exception as e:
             logger.bind(tag=TAG).error(f"发送TTS文本失败: {str(e)}")
@@ -302,7 +301,7 @@ class TTSProvider(TTSProviderBase):
                 event=EVENT_StartSession, sessionId=session_id
             ).as_bytes()
             payload = self.get_payload_bytes(
-                event=EVENT_StartSession, speaker=self.speaker
+                event=EVENT_StartSession, speaker=self.voice
             )
             await self.send_event(header, optional, payload)
             logger.bind(tag=TAG).info("会话启动请求已发送")

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

@@ -14,9 +14,9 @@ class TTSProvider(TTSProviderBase):
         self.api_key = config.get("api_key")
         self.model = config.get("model")
         if config.get("private_voice"):
-            self.voice_id = config.get("private_voice")
+            self.voice = config.get("private_voice")
         else:
-            self.voice_id = config.get("voice_id")
+            self.voice = config.get("voice_id")
 
         default_voice_setting = {
             "voice_id": "female-shaonv",
@@ -43,8 +43,8 @@ class TTSProvider(TTSProviderBase):
         self.audio_setting = {**defult_audio_setting, **config.get("audio_setting", {})}
         self.timber_weights = parse_string_to_list(config.get("timber_weights"))
 
-        if self.voice_id:
-            self.voice_setting["voice_id"] = self.voice_id
+        if self.voice:
+            self.voice_setting["voice_id"] = self.voice
 
         self.host = "api.minimax.chat"
         self.api_url = f"https://{self.host}/v1/t2a_v2?GroupId={self.group_id}"

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

@@ -19,9 +19,9 @@ class TTSProvider(TTSProviderBase):
             "https://u95167-bd74-2aef8085.westx.seetacloud.com:8443/flashsummary/tts?token=",
         )
         if config.get("private_voice"):
-            self.voice_id = int(config.get("private_voice"))
+            self.voice = int(config.get("private_voice"))
         else:
-            self.voice_id = int(config.get("voice_id", 1695))
+            self.voice = int(config.get("voice_id", 1695))
         self.token = config.get("token")
         self.to_lang = config.get("to_lang")
         self.volume_change_dB = int(config.get("volume_change_dB", 0))
@@ -50,7 +50,7 @@ class TTSProvider(TTSProviderBase):
                 "emotion": self.emotion,
                 "format": self.format,
                 "volume_change_dB": self.volume_change_dB,
-                "voice_id": self.voice_id,
+                "voice_id": self.voice,
                 "pitch_factor": self.pitch_factor,
                 "speed_factor": self.speed_factor,
                 "token": self.token,

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

@@ -0,0 +1,143 @@
+import os
+import yaml
+import time
+import hashlib
+import fcntl
+from typing import Dict
+from contextlib import contextmanager
+
+
+class FileLock:
+    def __init__(self, file, timeout=5):
+        self.file = file
+        self.timeout = timeout
+        self.start_time = None
+
+    def __enter__(self):
+        self.start_time = time.time()
+        while True:
+            try:
+                fcntl.flock(self.file, fcntl.LOCK_EX | fcntl.LOCK_NB)
+                return self.file
+            except IOError:
+                if time.time() - self.start_time > self.timeout:
+                    raise TimeoutError("获取文件锁超时")
+                time.sleep(0.1)
+
+    def __exit__(self, exc_type, exc_val, exc_tb):
+        fcntl.flock(self.file, fcntl.LOCK_UN)
+
+
+class WakeupWordsConfig:
+    def __init__(self):
+        self.config_file = "data/.wakeup_words.yaml"
+        self.assets_dir = "config/assets/wakeup_words"
+        self._ensure_directories()
+        self._config_cache = None
+        self._last_load_time = 0
+        self._cache_ttl = 1  # 缓存有效期（秒）
+        self._lock_timeout = 5  # 文件锁超时时间（秒）
+
+    def _ensure_directories(self):
+        """确保必要的目录存在"""
+        os.makedirs(os.path.dirname(self.config_file), exist_ok=True)
+        os.makedirs(self.assets_dir, exist_ok=True)
+
+    def _load_config(self) -> Dict:
+        """加载配置文件，使用缓存机制"""
+        current_time = time.time()
+
+        # 如果缓存有效，直接返回缓存
+        if (
+            self._config_cache is not None
+            and current_time - self._last_load_time < self._cache_ttl
+        ):
+            return self._config_cache
+
+        try:
+            with open(self.config_file, "a+") as f:
+                with FileLock(f, timeout=self._lock_timeout):
+                    f.seek(0)
+                    content = f.read()
+                    config = yaml.safe_load(content) if content else {}
+                    self._config_cache = config
+                    self._last_load_time = current_time
+                    return config
+        except (TimeoutError, IOError) as e:
+            print(f"加载配置文件失败: {e}")
+            return {}
+        except Exception as e:
+            print(f"加载配置文件时发生未知错误: {e}")
+            return {}
+
+    def _save_config(self, config: Dict):
+        """保存配置到文件，使用文件锁保护"""
+        try:
+            with open(self.config_file, "w") as f:
+                with FileLock(f, timeout=self._lock_timeout):
+                    yaml.dump(config, f, allow_unicode=True)
+                    self._config_cache = config
+                    self._last_load_time = time.time()
+        except (TimeoutError, IOError) as e:
+            print(f"保存配置文件失败: {e}")
+            raise
+        except Exception as e:
+            print(f"保存配置文件时发生未知错误: {e}")
+            raise
+
+    def get_wakeup_response(self, voice: str) -> Dict:
+        voice = hashlib.md5(voice.encode()).hexdigest()
+        """获取唤醒词回复配置"""
+        config = self._load_config()
+        default_response = {
+            "voice": "default",
+            "file_path": "config/assets/wakeup_words.wav",
+            "time": 0,
+            "text": "哈啰啊，我是小智啦，声音好听的台湾女孩一枚，超开心认识你耶，最近在忙啥，别忘了给我来点有趣的料哦，我超爱听八卦的啦",
+        }
+
+        if not config or voice not in config:
+            return default_response
+
+        # 检查文件大小
+        file_path = config[voice]["file_path"]
+        if not os.path.exists(file_path) or os.stat(file_path).st_size < (15 * 1024):
+            return default_response
+
+        return config[voice]
+
+    def update_wakeup_response(self, voice: str, file_path: str, text: str):
+        """更新唤醒词回复配置"""
+        try:
+            config = self._load_config()
+            voice_hash = hashlib.md5(voice.encode()).hexdigest()
+            config[voice_hash] = {
+                "voice": voice,
+                "file_path": file_path,
+                "time": time.time(),
+                "text": text,
+            }
+            self._save_config(config)
+        except Exception as e:
+            print(f"更新唤醒词回复配置失败: {e}")
+            raise
+
+    def generate_file_path(self, voice: str) -> str:
+        """生成音频文件路径，使用voice的哈希值作为文件名"""
+        try:
+            # 生成voice的哈希值
+            voice_hash = hashlib.md5(voice.encode()).hexdigest()
+            file_path = os.path.join(self.assets_dir, f"{voice_hash}.wav")
+
+            # 如果文件已存在，先删除
+            if os.path.exists(file_path):
+                try:
+                    os.remove(file_path)
+                except Exception as e:
+                    print(f"删除已存在的音频文件失败: {e}")
+                    raise
+
+            return file_path
+        except Exception as e:
+            print(f"生成音频文件路径失败: {e}")
+            raise