update:豆包tts流式ASR空字符串问题

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

@@ -30,8 +30,8 @@ async def handleAudioMessage(conn, audio):
 
 
 async def resume_vad_detection(conn):
-    # 等待1秒后恢复VAD检测
-    await asyncio.sleep(1)
+    # 等待2秒后恢复VAD检测
+    await asyncio.sleep(2)
     conn.just_woken_up = False
 
 

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

@@ -82,9 +82,9 @@ class ASRProviderBase(ABC):
         )  # 确保ASR模块返回原始文本
         conn.logger.bind(tag=TAG).info(f"识别文本: {raw_text}")
         text_len, _ = remove_punctuation_and_length(raw_text)
+        self.stop_ws_connection()
         if text_len > 0:
             # 使用自定义模块进行上报
-            self.stop_ws_connection()
             await startToChat(conn, raw_text)
             enqueue_asr_report(conn, raw_text, asr_audio_task)
 

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

@@ -54,12 +54,8 @@ class ASRProvider(ASRProviderBase):
         await super().open_audio_channels(conn)
 
     async def receive_audio(self, conn, audio, audio_have_voice):
-        if not isinstance(audio, bytes):
-            return
-
         conn.asr_audio.append(audio)
-        if audio_have_voice == False and conn.client_have_voice == False:
-            conn.asr_audio = conn.asr_audio[-10:]
+        conn.asr_audio = conn.asr_audio[-10:]
 
         # 如果本次有声音，且之前没有建立连接
         if audio_have_voice and self.asr_ws is None and not self.is_processing:
@@ -114,9 +110,6 @@ class ASRProvider(ASRProviderBase):
 
                 # 发送缓存的音频数据
                 if conn.asr_audio and len(conn.asr_audio) > 0:
-                    logger.bind(tag=TAG).info(
-                        f"发送缓存音频数据: {len(conn.asr_audio)}"
-                    )
                     for cached_audio in conn.asr_audio[-10:]:
                         try:
                             pcm_frame = self.decoder.decode(cached_audio, 960)
@@ -166,6 +159,18 @@ class ASRProvider(ASRProviderBase):
                         payload = result["payload_msg"]
                         if "result" in payload:
                             utterances = payload["result"].get("utterances", [])
+                            # 检查duration和空文本的情况
+                            if (
+                                payload.get("audio_info", {}).get("duration", 0) > 2000
+                                and not utterances
+                                and not payload["result"].get("text")
+                            ):
+                                logger.bind(tag=TAG).error(f"识别文本：空")
+                                self.text = ""
+                                conn.reset_vad_states()
+                                await self.handle_voice_stop(conn, None)
+                                break
+
                             for utterance in utterances:
                                 if utterance.get("definite", False):
                                     self.text = utterance["text"]
@@ -235,7 +240,7 @@ class ASRProvider(ASRProviderBase):
                 "sample_rate": self.rate,
             },
         }
-        logger.bind(tag=TAG).info(
+        logger.bind(tag=TAG).debug(
             f"构造请求参数: {json.dumps(req, ensure_ascii=False)}"
         )
         return req

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

@@ -195,7 +195,7 @@ class TTSProvider(TTSProviderBase):
             try:
                 logger.bind(tag=TAG).debug("等待TTS文本队列消息...")
                 message = self.tts_text_queue.get(timeout=1)
-                logger.bind(tag=TAG).info(
+                logger.bind(tag=TAG).debug(
                     f"收到TTS任务｜{message.sentence_type.name} ｜ {message.content_type.name} | 会话ID: {self.conn.sentence_id}"
                 )
                 if self.conn.client_abort:
@@ -221,7 +221,7 @@ class TTSProvider(TTSProviderBase):
                 elif ContentType.TEXT == message.content_type:
                     if message.content_detail:
                         try:
-                            logger.bind(tag=TAG).info(
+                            logger.bind(tag=TAG).debug(
                                 f"开始发送TTS文本: {message.content_detail}"
                             )
                             future = asyncio.run_coroutine_threadsafe(
@@ -229,7 +229,7 @@ class TTSProvider(TTSProviderBase):
                                 loop=self.conn.loop,
                             )
                             future.result()
-                            logger.bind(tag=TAG).info("TTS文本发送成功")
+                            logger.bind(tag=TAG).debug("TTS文本发送成功")
                         except Exception as e:
                             logger.bind(tag=TAG).error(f"发送TTS文本失败: {str(e)}")
                             continue
@@ -250,7 +250,6 @@ class TTSProvider(TTSProviderBase):
                             loop=self.conn.loop,
                         )
                         future.result()
-                        logger.bind(tag=TAG).info("TTS会话结束成功")
                     except Exception as e:
                         logger.bind(tag=TAG).error(f"结束TTS会话失败: {str(e)}")
                         continue