Merge pull request #629 from joey-zhou/vad_optimize

优化vad，采用bytearray替换byte，提高字符串拼接效率

main/xiaozhi-server/core/connection.py

@@ -73,7 +73,7 @@ class ConnectionHandler:
         self.intent = _intent
 
         # vad相关变量
-        self.client_audio_buffer = bytes()
+        self.client_audio_buffer = bytearray()
         self.client_have_voice = False
         self.client_have_voice_last_time = 0.0
         self.client_no_voice_last_time = 0.0
@@ -804,7 +804,7 @@ class ConnectionHandler:
             # q.queue.put(None)
 
     def reset_vad_states(self):
-        self.client_audio_buffer = bytes()
+        self.client_audio_buffer = bytearray()
         self.client_have_voice = False
         self.client_have_voice_last_time = 0
         self.client_voice_stop = False

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

@@ -31,7 +31,7 @@ class SileroVAD(VAD):
     def is_vad(self, conn, opus_packet):
         try:
             pcm_frame = self.decoder.decode(opus_packet, 960)
-            conn.client_audio_buffer += pcm_frame  # 将新数据加入缓冲区
+            conn.client_audio_buffer.extend(pcm_frame)  # 将新数据加入缓冲区
 
             # 处理缓冲区中的完整帧（每次处理512采样点）
             client_have_voice = False