Merge pull request #2357 from kkroid/feature_tts-audio-send-delay

feat: 添加可配置的TTS音频发送延迟功能

main/xiaozhi-server/config.yaml

@@ -70,6 +70,12 @@ enable_stop_tts_notify: false
 # 说完话是否开启提示音，音效地址
 stop_tts_notify_voice: "config/assets/tts_notify.mp3"
 
+# TTS音频发送延迟配置
+# tts_audio_send_delay: 控制音频包发送间隔
+#   0: 使用精确时间控制，严格匹配音频帧率（默认，运行时按音频帧率计算）
+#   > 0: 使用固定延迟（毫秒）发送，例如: 60
+tts_audio_send_delay: 0
+
 exit_commands:
   - "退出"
   - "关闭"
@@ -987,4 +993,4 @@ TTS:
     # sample_rate: 24000  # 采样率：16000, 8000, 24000
     # volume: 50  # 音量：0-100
     # speed: 50  # 语速：0-100
-    # pitch: 50  # 语调：0-100
+    # pitch: 50  # 语调：0-100

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

@@ -90,6 +90,9 @@ async def sendAudio(conn, audios, frame_duration=60):
     if audios is None or len(audios) == 0:
         return
 
+    # 获取发送延迟配置
+    send_delay = conn.config.get("tts_audio_send_delay", -1) / 1000.0
+
     if isinstance(audios, bytes):
         if conn.client_abort:
             return
@@ -107,16 +110,21 @@ async def sendAudio(conn, audios, frame_duration=60):
 
         flow_control = conn.audio_flow_control
         current_time = time.perf_counter()
-        # 计算预期发送时间
-        expected_time = flow_control["start_time"] + (
-            flow_control["packet_count"] * frame_duration / 1000
-        )
-        delay = expected_time - current_time
-        if delay > 0:
-            await asyncio.sleep(delay)
+        
+        if send_delay > 0:
+            # 使用固定延迟
+            await asyncio.sleep(send_delay)
         else:
-            # 纠正误差
-            flow_control["start_time"] += abs(delay)
+            # 计算预期发送时间
+            expected_time = flow_control["start_time"] + (
+                flow_control["packet_count"] * frame_duration / 1000
+            )
+            delay = expected_time - current_time
+            if delay > 0:
+                await asyncio.sleep(delay)
+            else:
+                # 纠正误差
+                flow_control["start_time"] += abs(delay)
 
         if conn.conn_from_mqtt_gateway:
             # 计算时间戳和序列号
@@ -164,12 +172,16 @@ async def sendAudio(conn, audios, frame_duration=60):
             # 重置没有声音的状态
             conn.last_activity_time = time.time() * 1000
 
-            # 计算预期发送时间
-            expected_time = start_time + (play_position / 1000)
-            current_time = time.perf_counter()
-            delay = expected_time - current_time
-            if delay > 0:
-                await asyncio.sleep(delay)
+            if send_delay > 0:
+                # 固定延迟模式
+                await asyncio.sleep(send_delay)
+            else:
+                 # 计算预期发送时间
+                expected_time = start_time + (play_position / 1000)
+                current_time = time.perf_counter()
+                delay = expected_time - current_time
+                if delay > 0:
+                    await asyncio.sleep(delay)
 
             if conn.conn_from_mqtt_gateway:
                 # 计算时间戳和序列号（使用当前的数据包索引确保连续性）