a little cleanup

src/samples/deprecated/static-sprite/sprite-sample.py

@@ -40,7 +40,7 @@ class StaticSpriteResponse(OrchestratorResponse):
             self.image_bytes = img.tobytes()
 
     def do_play(self) -> None:
-        self.output_queue.put(QueueFrame(FrameType.IMAGE_FRAME, self.image_bytes))
+        self.output_queue.put(QueueFrame(FrameType.IMAGE, self.image_bytes))
 
 
 class IntroSpriteResponse(StaticSpriteResponse):
@@ -73,8 +73,8 @@ class AnimatedSpriteLLMResponse(LLMResponse):
 
     def get_frames_from_tts_response(self, audio_frame) -> list[QueueFrame]:
         return [
-            QueueFrame(FrameType.AUDIO_FRAME, audio_frame),
-            QueueFrame(FrameType.IMAGE_FRAME, random.choice(self.image_bytes))
+            QueueFrame(FrameType.AUDIO, audio_frame),
+            QueueFrame(FrameType.IMAGE, random.choice(self.image_bytes))
         ]
 
 

src/samples/theoretical-to-real/01-say-one-thing.py

@@ -38,7 +38,7 @@ async def main(room_url):
             return
 
         async for audio in audio_generator:
-            transport.output_queue.put(QueueFrame(FrameType.AUDIO_FRAME, audio))
+            transport.output_queue.put(QueueFrame(FrameType.AUDIO, audio))
 
         # wait for the output queue to be empty, then leave the meeting
         transport.output_queue.join()

src/samples/theoretical-to-real/01a-greet-user.py

@@ -41,7 +41,7 @@ async def main(room_url):
         audio_generator: AsyncGenerator[bytes, None] = tts.run_tts(f"Hello there, {participant['info']['userName']}!")
 
         async for audio in audio_generator:
-            transport.output_queue.put(QueueFrame(FrameType.AUDIO_FRAME, audio))
+            transport.output_queue.put(QueueFrame(FrameType.AUDIO, audio))
 
     print("setting up call state handler")
     @transport.event_handler("on_call_state_updated")

src/samples/theoretical-to-real/02-llm-say-one-thing.py

@@ -29,7 +29,7 @@ async def main(room_url):
         "role": "system",
         "content": "You are an LLM in a WebRTC session, and this is a 'hello world' demo. Say hello to the world."
     }]
-    await text_to_llm_queue.put(QueueFrame(FrameType.LLM_MESSAGE_FRAME, messages))
+    await text_to_llm_queue.put(QueueFrame(FrameType.LLM_MESSAGE, messages))
     await text_to_llm_queue.put(QueueFrame(FrameType.END_STREAM, None))
 
     llm_task = asyncio.create_task(llm.run())

src/samples/theoretical-to-real/03-still-frame.py

@@ -27,7 +27,7 @@ async def main(room_url):
     @transport.event_handler("on_participant_joined")
     async def on_participant_joined(transport, participant):
         (_, image_bytes) = await image_task
-        transport.output_queue.put(QueueFrame(FrameType.IMAGE_FRAME, image_bytes))
+        transport.output_queue.put(QueueFrame(FrameType.IMAGE, image_bytes))
 
     await transport.run()
 

src/samples/theoretical-to-real/04-utterance-and-speech.py

@@ -41,11 +41,11 @@ async def main(room_url:str):
         ))
 
         async for audio_chunk in tts.run_tts("My friend the LLM is now going to tell a joke about llamas."):
-            transport.output_queue.put(QueueFrame(FrameType.AUDIO_FRAME, audio_chunk))
+            transport.output_queue.put(QueueFrame(FrameType.AUDIO, audio_chunk))
 
         llm_response = await llm_response_task
         async for audio_chunk in tts.run_tts(llm_response):
-            transport.output_queue.put(QueueFrame(FrameType.AUDIO_FRAME, audio_chunk))
+            transport.output_queue.put(QueueFrame(FrameType.AUDIO, audio_chunk))
 
 
         # wait for the output queue to be empty, then leave the meeting

src/samples/theoretical-to-real/05-sync-speech-and-text.py

@@ -98,12 +98,12 @@ async def main(room_url):
             data = await month_data_task
             transport.output_queue.put(
                 [
-                    QueueFrame(FrameType.IMAGE_FRAME, data["image"]),
-                    QueueFrame(FrameType.AUDIO_FRAME, data["audio"][0]),
+                    QueueFrame(FrameType.IMAGE, data["image"]),
+                    QueueFrame(FrameType.AUDIO, data["audio"][0]),
                 ]
             )
             for audio in data["audio"][1:]:
-                transport.output_queue.put(QueueFrame(FrameType.AUDIO_FRAME, audio))
+                transport.output_queue.put(QueueFrame(FrameType.AUDIO, audio))
 
         # wait for the output queue to be empty, then leave the meeting
         transport.output_queue.join()

src/samples/theoretical-to-real/06-listen-and-respond.py

@@ -32,7 +32,11 @@ async def main(room_url:str, token):
         ]
 
         sentence = ""
-        async for message in transport.get_transcriptions():
+        async for frame in transport.get_media_frames():
+            if frame.frame_type != FrameType.TRANSCRIPTION:
+                continue
+
+            message = frame.frame_data
             if message["session_id"] == transport.my_participant_id:
                 continue
 
@@ -46,7 +50,7 @@ async def main(room_url:str, token):
                 async for response in llm.run_llm_async_sentences(messages):
                     full_response += response
                     async for audio in tts.run_tts(response):
-                        transport.output_queue.put(QueueFrame(FrameType.AUDIO_FRAME, audio))
+                        await transport.output_queue.put(QueueFrame(FrameType.AUDIO, audio))
 
                 messages.append({"role": "assistant", "content": full_response})