语音发送新增同步机制

jiangtao · jiangtao · commit 9387a5a5667e · 2025-02-26T22:34:28.000+08:00
diff --git a/core/providers/tts/base.py b/core/providers/tts/base.py
@@ -52,28 +52,25 @@ def get_opus_data(self, file_path):
             # 获取音频时长
             duration = self.get_audio_duration(file_path)
             
-            # 初始化Opus编码器（用于验证帧大小）
-            encoder = opuslib_next.Encoder(16000, 1, opuslib_next.APPLICATION_AUDIO)
-            
-            # 编码参数（与原wav转换保持一致）
-            frame_duration = 60  # 60ms per frame
-            frame_size = int(16000 * frame_duration / 1000)  # 960 samples/frame
-            
             opus_datas = []
             current_pos = 0
             
             while current_pos < len(opus_data):
-                # 读取帧数据
-                frame_data = opus_data[current_pos:current_pos + frame_size]
-                if len(frame_data) < frame_size:
-                    # 如果最后一帧不足，补零
-                    frame_data += b'\x00' * (frame_size - len(frame_data))
+                # 读取帧长度（前2个字节）
+                if current_pos + 2 > len(opus_data):
+                    break
+                    
+                frame_length = int.from_bytes(opus_data[current_pos:current_pos + 2], 'little')
+                current_pos += 2
                 
-                # 使用与原方法相同的编码方式
-                encoded_data = encoder.encode(frame_data, frame_size)
-                opus_datas.append(encoded_data)
-                
-                current_pos += frame_size
+                # 确保有足够的数据读取
+                if current_pos + frame_length > len(opus_data):
+                    break
+                    
+                # 读取帧数据
+                frame_data = opus_data[current_pos:current_pos + frame_length]
+                opus_datas.append(frame_data)
+                current_pos += frame_length
             
             return opus_datas, duration
             
diff --git a/core/providers/tts/doubao.py b/core/providers/tts/doubao.py
@@ -39,7 +39,7 @@ async def text_to_speak(self, text, output_file):
                 "speed_ratio": 1.0,
                 "volume_ratio": 1.0,
                 "pitch_ratio": 1.0,
-                "rate": 16000,
+                "rate": 16000
             },
             "request": {
                 "reqid": str(uuid.uuid4()),
