services: fixed DailyTransport output camera and audio

CHANGELOG.md

@@ -15,6 +15,9 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### Fixed
 
+- Fixed an issue with `DailyInputTransport` and `DailyOutputTransport` that
+  could cause some threads to not start properly.
+
 - Fixed `STTService`. Add `max_silence_secs` and `max_buffer_secs` to handle
   better what's being passed to the STT service. Also add exponential smoothing
   to the RMS.

src/pipecat/transports/services/daily.py

@@ -416,23 +416,18 @@ class DailyInputTransport(BaseInputTransport):
         self._camera_in_thread = threading.Thread(target=self._camera_in_thread_handler)
 
     async def start(self):
-        if self._running:
-            return
-
+        await super().start()
         self._camera_in_thread.start()
         await self._session.join()
-        await super().start()
 
     async def stop(self):
-        await self._session.leave()
         await super().stop()
+        await self._session.leave()
 
     async def cleanup(self):
-        self._camera_in_thread.join()
-
-        await self._session.cleanup()
-
         await super().cleanup()
+        self._camera_in_thread.join()
+        await self._session.cleanup()
 
     def vad_analyze(self, audio_frames: bytes) -> VADState:
         return self._session.vad_analyze(audio_frames)
@@ -523,19 +518,16 @@ class DailyOutputTransport(BaseOutputTransport):
         self._session = session
 
     async def start(self):
-        if self._running:
-            return
-
-        await self._session.join()
         await super().start()
+        await self._session.join()
 
     async def stop(self):
-        await self._session.leave()
         await super().stop()
+        await self._session.leave()
 
     async def cleanup(self):
-        await self._session.cleanup()
         await super().cleanup()
+        await self._session.cleanup()
 
     def write_raw_audio_frames(self, frames: bytes):
         self._session.write_raw_audio_frames(frames)