updated

bot/bot_websocket_server.py

@@ -66,128 +66,100 @@ load_dotenv(override=True)
 
 
 async def run_bot_websocket_server(websocket_client):
-    ws_transport = FastAPIWebsocketTransport(
-        websocket=websocket_client,
-        params=FastAPIWebsocketParams(
-            audio_in_enabled=True,
-            audio_out_enabled=True,
-            add_wav_header=False,
-            vad_analyzer=SileroVADAnalyzer(),
-            serializer=ProtobufFrameSerializer(),
-        ),
-    )
-
-    stt = WhisperSTTService(
-        model="tiny",
-        device="cpu",
-        compute_type="default",
-        language="en",
-    )
-
-    llm = OLLamaLLMService(
-        model="smollm:latest",
-        # params=OLLamaLLMService.InputParams(temperature=0.7, max_tokens=1000),
-    )
-
-    # TTS = FishAudioTTSService(
-    #     api_key=os.getenv("CARTESIA_API_KEY"),
-    #     voice_id="79a125e8-cd45-4c13-8a67-188112f4dd22",  # British Reading Lady
-    # )
-    # async with aiohttp.ClientSession() as session:
-    #     TTS = XTTSService(
-    #         voice_id="speaker_1",
-    #         language=Language.EN,
-    #         base_url="http://localhost:8000",
-    #         aiohttp_session=session
-    #     )
-
-    context = OpenAILLMContext(
-        [
-            {
-                "role": "system",
-                "content": SYSTEM_INSTRUCTION,
-            },
-            {
-                "role": "user",
-                "content": "Start by greeting the user warmly and introducing yourself.",
-            },
-        ],
-    )
-    context_aggregator = llm.create_context_aggregator(context)
-
-    # RTVI events for Pipecat client UI
-    rtvi = RTVIProcessor(config=RTVIConfig(config=[]))
-
-    TTS = KokoroTTSService(
-        model_path=os.path.join(
-            os.path.dirname(__file__), "assets", "kokoro-v1.0.int8.onnx"
-        ),
-        voices_path=os.path.join(os.path.dirname(__file__), "assets", "voices.json"),
-        voice_id="af",
-        sample_rate=16000,
-    )
-
-    # TTS = OrpheusTTSService(
-    #     model_name="canopylabs/orpheus-3b-0.1-ft",
-    #     sample_rate=16000,
-    # )
-
-    # TTS = ChatterboxTTSService(
-    #     model_name="",
-    #     sample_rate=16000,
-    # )
-
-    # TTS = DiaTTSService(
-    #     model_name="nari-labs/Dia-1.6B",
-    #     sample_rate=16000,
-    # )
-    pipeline = Pipeline(
-        [
-            ws_transport.input(),
-            rtvi,
-            stt,  # STT
-            context_aggregator.user(),
-            llm,
-            TTS,  # TTS
-            ws_transport.output(),
-            context_aggregator.assistant(),
-        ]
-    )
-
-    task = PipelineTask(
-        pipeline,
-        params=PipelineParams(
-            enable_metrics=True,
-            allow_interruptions=True,
-            enable_usage_metrics=True,
-        ),
-        # enable_turn_tracking=True,
-        enable_tracing=False,
-        conversation_id="test",
-        observers=[RTVIObserver(rtvi)],
-    )
-
-    @rtvi.event_handler("on_client_ready")
-    async def on_client_ready(rtvi):
-        logger.info("Pipecat client ready.")
-        await rtvi.set_bot_ready()
-        # Kick off the conversation.
-        await task.queue_frames([context_aggregator.user().get_context_frame()])
-
-    @ws_transport.event_handler("on_client_connected")
-    async def on_client_connected(transport, client):
-        logger.info("Pipecat Client connected")
-
-    @ws_transport.event_handler("on_client_disconnected")
-    async def on_client_disconnected(transport, client):
-        logger.info("Pipecat Client disconnected")
-        await task.cancel()
-
-    @ws_transport.event_handler("on_session_timeout")
-    async def on_session_timeout(transport, client):
-        logger.info(f"Entering in timeout for {client.remote_address}")
-        await task.cancel()
-
-    runner = PipelineRunner()
-
-    await runner.run(task)
+    try:
+        ws_transport = FastAPIWebsocketTransport(
+            websocket=websocket_client,
+            params=FastAPIWebsocketParams(
+                audio_in_enabled=True,
+                audio_out_enabled=True,
+                add_wav_header=False,
+                vad_analyzer=SileroVADAnalyzer(),
+                serializer=ProtobufFrameSerializer(),
+            ),
+        )
+
+        stt = WhisperSTTService(
+            model="tiny",
+            device="cpu",
+            compute_type="default",
+            language="en",
+        )
+
+        llm = OLLamaLLMService(
+            model="smollm:latest",
+        )
+
+        context = OpenAILLMContext(
+            [
+                {"role": "system", "content": SYSTEM_INSTRUCTION},
+                {
+                    "role": "user",
+                    "content": "Start by greeting the user warmly and introducing yourself.",
+                },
+            ]
+        )
+        context_aggregator = llm.create_context_aggregator(context)
+
+        rtvi = RTVIProcessor(config=RTVIConfig(config=[]))
+
+        TTS = KokoroTTSService(
+            model_path=os.path.join(
+                os.path.dirname(__file__), "assets", "kokoro-v1.0.int8.onnx"
+            ),
+            voices_path=os.path.join(
+                os.path.dirname(__file__), "assets", "voices.json"
+            ),
+            voice_id="af",
+            sample_rate=16000,
+        )
+
+        pipeline = Pipeline(
+            [
+                ws_transport.input(),
+                rtvi,
+                stt,
+                context_aggregator.user(),
+                llm,
+                TTS,
+                ws_transport.output(),
+                context_aggregator.assistant(),
+            ]
+        )
+
+        task = PipelineTask(
+            pipeline,
+            params=PipelineParams(
+                enable_metrics=True,
+                allow_interruptions=True,
+                enable_usage_metrics=True,
+            ),
+            enable_tracing=False,
+            conversation_id="test",
+            observers=[RTVIObserver(rtvi)],
+        )
+
+        @rtvi.event_handler("on_client_ready")
+        async def on_client_ready(rtvi):
+            logger.info("Pipecat client ready.")
+            await rtvi.set_bot_ready()
+            await task.queue_frames([context_aggregator.user().get_context_frame()])
+
+        @ws_transport.event_handler("on_client_connected")
+        async def on_client_connected(transport, client):
+            logger.info("Pipecat Client connected")
+
+        @ws_transport.event_handler("on_client_disconnected")
+        async def on_client_disconnected(transport, client):
+            logger.info("Pipecat Client disconnected")
+            await task.cancel()
+
+        @ws_transport.event_handler("on_session_timeout")
+        async def on_session_timeout(transport, client):
+            logger.info(f"Entering in timeout for {client.remote_address}")
+            await task.cancel()
+
+        runner = PipelineRunner()
+        await runner.run(task)
+
+    except Exception as e:
+        logger.exception("Error in run_bot_websocket_server")