debug: 增加了讯飞语音识别接口的timeout机制，防止语音识别未完成时连接断开

.gitignore

@@ -12,3 +12,6 @@ nohup.out
 
 /app/takway-ai.top.key
 /app/takway-ai.top.pem
+
+tests/assets/BarbieDollsVoice.mp3
+utils/tts/openvoice_model/checkpoint.pth

app/controllers/chat_controller.py

@@ -8,7 +8,7 @@ from ..models import UserCharacter, Session, Character, User, Audio
 from utils.audio_utils import VAD
 from fastapi import WebSocket, HTTPException, status
 from datetime import datetime
-from utils.xf_asr_utils import generate_xf_asr_url, make_first_frame, make_continue_frame, make_last_frame, parse_xfasr_recv
+from utils.xf_asr_utils import xf_asr_websocket_factory, make_first_frame, make_continue_frame, make_last_frame, parse_xfasr_recv
 from config import get_config
 import numpy as np
 import websockets
@@ -280,23 +280,42 @@ async def sct_asr_handler(ws,session_id,user_input_q,llm_input_q,user_input_fini
         logger.debug(f"接收到用户消息: {current_message}")
     elif Config.STRAM_CHAT.ASR == "XF":
         status = FIRST_FRAME
-        async with websockets.connect(generate_xf_asr_url()) as xf_websocket:
+        xf_websocket = await xf_asr_websocket_factory() #获取一个讯飞语音识别接口websocket连接
+        segment_duration_threshold = 25  #设置一个连接时长上限，讯飞语音接口超过30秒会自动断开连接，所以该值设置成25秒
+        segment_start_time = asyncio.get_event_loop().time()
+        current_message = ""
         while not (user_input_finish_event.is_set() and user_input_q.empty()):
+            try:
                 audio_data = await user_input_q.get()
+                current_time = asyncio.get_event_loop().time()
+                if current_time - segment_start_time > segment_duration_threshold:
+                    await xf_websocket.send(make_last_frame())
+                    current_message += parse_xfasr_recv(await xf_websocket.recv())
+                    await xf_websocket.close()
+                    xf_websocket = await xf_asr_websocket_factory() #重建一个websocket连接
+                    status = FIRST_FRAME
+                    segment_start_time = current_time
                 if status == FIRST_FRAME:
                     await xf_websocket.send(make_first_frame(audio_data))
                     status = CONTINUE_FRAME
                 elif status == CONTINUE_FRAME:
                     await xf_websocket.send(make_continue_frame(audio_data))
-            await xf_websocket.send(make_last_frame(""))
+            except websockets.exceptions.ConnectionClosedOK:
-
+                logger.debug("讯飞语音识别接口连接断开,重新创建连接")
-            current_message = parse_xfasr_recv(json.loads(await xf_websocket.recv()))
+                xf_websocket = await xf_asr_websocket_factory() #重建一个websocket连接
-            if current_message in ["","嗯"]:
+                status = FIRST_FRAME
+                segment_start_time = asyncio.get_event_loop().time()
+        await xf_websocket.send(make_last_frame())
+        current_message += parse_xfasr_recv(await xf_websocket.recv())
+        await xf_websocket.close()
+        if current_message in ["嗯", ""]:
             await ws.send_text(json.dumps({"type": "close", "code": 201, "msg": ""}, ensure_ascii=False))
             return
         await llm_input_q.put(current_message)
         logger.debug(f"接收到用户消息: {current_message}")
             
+                
+            
 #大模型调用
 async def sct_llm_handler(ws,session_id,response_type,llm_info,tts_info,db,redis,llm_input_q,chat_finished_event):
     logger.debug("llm调用函数启动")

utils/xf_asr_utils.py

@@ -1,3 +1,4 @@
+import websockets
 import datetime
 import hashlib
 import base64
@@ -62,3 +63,7 @@ def parse_xfasr_recv(message):
             for w in i['cw']:
                 result += w['w']
         return result
+    
+async def xf_asr_websocket_factory():
+    url = generate_xf_asr_url()
+    return await websockets.connect(url)