Merge pull request #4120 from pipecat-ai/filipi/krisp-viva-vad-support

Added cleanup() method to VADAnalyzer base class

changelog/4120.added.md

@@ -0,0 +1 @@
+- Added `cleanup()` method to `VADAnalyzer` and `VADController` so VAD analyzer resources are properly released when no longer needed. Custom `VADAnalyzer` subclasses can override `cleanup()` to free any held resources.

examples/foundational/07p-interruptible-krisp-viva.py

@@ -30,7 +30,6 @@ from loguru import logger
 from pipecat.audio.filters.krisp_viva_filter import KrispVivaFilter
 from pipecat.audio.turn.krisp_viva_turn import KrispVivaTurn
 from pipecat.audio.vad.silero import SileroVADAnalyzer
-from pipecat.audio.vad.vad_analyzer import VADParams
 from pipecat.frames.frames import LLMRunFrame
 from pipecat.metrics.metrics import TurnMetricsData
 from pipecat.observers.loggers.metrics_log_observer import MetricsLogObserver
@@ -64,20 +63,17 @@ transport_params = {
     "daily": lambda: DailyParams(
         audio_in_enabled=True,
         audio_out_enabled=True,
-        vad_analyzer=SileroVADAnalyzer(params=VADParams(stop_secs=0.2)),  # or KrispVivaVadAnalyzer
-        audio_in_filter=KrispVivaFilter(),
+        audio_in_filter=krisp_viva_filter,
     ),
     "twilio": lambda: FastAPIWebsocketParams(
         audio_in_enabled=True,
         audio_out_enabled=True,
-        vad_analyzer=SileroVADAnalyzer(params=VADParams(stop_secs=0.2)),  # or KrispVivaVadAnalyzer
-        audio_in_filter=KrispVivaFilter(),
+        audio_in_filter=krisp_viva_filter,
     ),
     "webrtc": lambda: TransportParams(
         audio_in_enabled=True,
         audio_out_enabled=True,
-        vad_analyzer=SileroVADAnalyzer(params=VADParams(stop_secs=0.2)),  # or KrispVivaVadAnalyzer
-        audio_in_filter=KrispVivaFilter(),
+        audio_in_filter=krisp_viva_filter,
     ),
 }
 
@@ -108,7 +104,7 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
             user_turn_strategies=UserTurnStrategies(
                 stop=[TurnAnalyzerUserTurnStopStrategy(turn_analyzer=KrispVivaTurn())]
             ),
-            vad_analyzer=SileroVADAnalyzer(),
+            vad_analyzer=SileroVADAnalyzer(),  # or KrispVivaVadAnalyzer
         ),
     )
 

src/pipecat/audio/vad/vad_analyzer.py

@@ -242,3 +242,12 @@ class VADAnalyzer(ABC):
             self._vad_stopping_count = 0
 
         return self._vad_state
+
+    async def cleanup(self):
+        """Clean up resources.
+
+        This method should be called when the object is no longer needed.
+        It waits for all currently executing event handler tasks to finish
+        before returning.
+        """
+        pass

src/pipecat/audio/vad/vad_controller.py

@@ -109,6 +109,16 @@ class VADController(BaseObject):
         # Broadcast initial VAD params so other services (e.g. STT) can use them
         await self.broadcast_frame(SpeechControlParamsFrame, vad_params=self._vad_analyzer.params)
 
+    async def cleanup(self):
+        """Clean up resources.
+
+        This method should be called when the object is no longer needed.
+        It waits for all currently executing event handler tasks to finish
+        before returning.
+        """
+        if self._vad_analyzer:
+            await self._vad_analyzer.cleanup()
+
     async def _handle_audio(self, frame: InputAudioRawFrame):
         """Process an audio chunk and emit speech events as needed.