From f2982b4b68a5d823596d79c4119394e4c06a2401 Mon Sep 17 00:00:00 2001 From: vipyne Date: Wed, 26 Nov 2025 13:39:36 -0600 Subject: [PATCH] add observer to trace basic STT, LLM, & TTS spans --- .../foundational/30a-stt-llm-tts-observer.py | 147 +++++++++++++++++ .../observers/stt_llm_tts_trace_observer.py | 155 ++++++++++++++++++ 2 files changed, 302 insertions(+) create mode 100644 examples/foundational/30a-stt-llm-tts-observer.py create mode 100644 src/pipecat/observers/stt_llm_tts_trace_observer.py diff --git a/examples/foundational/30a-stt-llm-tts-observer.py b/examples/foundational/30a-stt-llm-tts-observer.py new file mode 100644 index 000000000..2dcf68326 --- /dev/null +++ b/examples/foundational/30a-stt-llm-tts-observer.py @@ -0,0 +1,147 @@ +# +# Copyright (c) 2024–2025, Daily +# +# SPDX-License-Identifier: BSD 2-Clause License +# + + +import os + +from dotenv import load_dotenv +from loguru import logger + +from pipecat.audio.turn.smart_turn.base_smart_turn import SmartTurnParams +from pipecat.audio.turn.smart_turn.local_smart_turn_v3 import LocalSmartTurnAnalyzerV3 +from pipecat.audio.vad.silero import SileroVADAnalyzer +from pipecat.audio.vad.vad_analyzer import VADParams +from pipecat.frames.frames import ( + BotStartedSpeakingFrame, + BotStoppedSpeakingFrame, + EndFrame, + InterruptionFrame, + LLMRunFrame, + TTSTextFrame, + UserStartedSpeakingFrame, +) +from pipecat.observers.base_observer import BaseObserver, FramePushed +from pipecat.observers.loggers.debug_log_observer import DebugLogObserver, FrameEndpoint +from pipecat.observers.loggers.transcription_log_observer import TranscriptionLogObserver +from pipecat.observers.stt_llm_tts_trace_observer import STTLLMTTSTraceObserver +from pipecat.pipeline.pipeline import Pipeline +from pipecat.pipeline.runner import PipelineRunner +from pipecat.pipeline.task import PipelineParams, PipelineTask +from pipecat.processors.aggregators.llm_context import LLMContext +from pipecat.processors.aggregators.llm_response_universal import LLMContextAggregatorPair +from pipecat.processors.frame_processor import FrameDirection +from pipecat.runner.types import RunnerArguments +from pipecat.runner.utils import create_transport +from pipecat.services.cartesia.tts import CartesiaTTSService +from pipecat.services.deepgram.stt import DeepgramSTTService +from pipecat.services.openai.llm import OpenAILLMService +from pipecat.transports.base_input import BaseInputTransport +from pipecat.transports.base_output import BaseOutputTransport +from pipecat.transports.base_transport import BaseTransport, TransportParams +from pipecat.transports.daily.transport import DailyParams +from pipecat.transports.websocket.fastapi import FastAPIWebsocketParams + +load_dotenv(override=True) + + +# We store functions so objects (e.g. SileroVADAnalyzer) don't get +# instantiated. The function will be called when the desired transport gets +# selected. +transport_params = { + "daily": lambda: DailyParams( + audio_in_enabled=True, + audio_out_enabled=True, + vad_analyzer=SileroVADAnalyzer(params=VADParams(stop_secs=0.2)), + turn_analyzer=LocalSmartTurnAnalyzerV3(params=SmartTurnParams()), + ), + "twilio": lambda: FastAPIWebsocketParams( + audio_in_enabled=True, + audio_out_enabled=True, + vad_analyzer=SileroVADAnalyzer(params=VADParams(stop_secs=0.2)), + turn_analyzer=LocalSmartTurnAnalyzerV3(params=SmartTurnParams()), + ), + "webrtc": lambda: TransportParams( + audio_in_enabled=True, + audio_out_enabled=True, + vad_analyzer=SileroVADAnalyzer(params=VADParams(stop_secs=0.2)), + turn_analyzer=LocalSmartTurnAnalyzerV3(params=SmartTurnParams()), + ), +} + + +async def run_bot(transport: BaseTransport, runner_args: RunnerArguments): + logger.info(f"Starting bot") + + stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY")) + + tts = CartesiaTTSService( + api_key=os.getenv("CARTESIA_API_KEY"), + voice_id="71a7ad14-091c-4e8e-a314-022ece01c121", # British Reading Lady + ) + + llm = OpenAILLMService(api_key=os.getenv("OPENAI_API_KEY")) + + messages = [ + { + "role": "system", + "content": "You are a helpful LLM in a WebRTC call. Your goal is to demonstrate your capabilities in a succinct way. Your output will be spoken aloud, so avoid special characters that can't easily be spoken, such as emojis or bullet points. Respond to what the user said in a creative and helpful way.", + }, + ] + + context = LLMContext(messages) + context_aggregator = LLMContextAggregatorPair(context) + + pipeline = Pipeline( + [ + transport.input(), # Transport user input + stt, + context_aggregator.user(), # User responses + llm, # LLM + tts, # TTS + transport.output(), # Transport bot output + context_aggregator.assistant(), # Assistant spoken responses + ] + ) + + task = PipelineTask( + pipeline, + params=PipelineParams( + enable_metrics=True, + enable_usage_metrics=True, + ), + idle_timeout_secs=runner_args.pipeline_idle_timeout_secs, + observers=[ + STTLLMTTSTraceObserver(), + ], + ) + + @transport.event_handler("on_client_connected") + async def on_client_connected(transport, client): + logger.info(f"Client connected") + # Kick off the conversation. + messages.append({"role": "system", "content": "Please introduce yourself to the user."}) + await task.queue_frames([LLMRunFrame()]) + + @transport.event_handler("on_client_disconnected") + async def on_client_disconnected(transport, client): + logger.info(f"Client disconnected") + await task.cancel() + + runner = PipelineRunner(handle_sigint=runner_args.handle_sigint) + + await runner.run(task) + + +async def bot(runner_args: RunnerArguments): + """Main bot entry point compatible with Pipecat Cloud.""" + transport = await create_transport(runner_args, transport_params) + await run_bot(transport, runner_args) + + +if __name__ == "__main__": + from pipecat.runner.run import main + + main() diff --git a/src/pipecat/observers/stt_llm_tts_trace_observer.py b/src/pipecat/observers/stt_llm_tts_trace_observer.py new file mode 100644 index 000000000..4bbff76ea --- /dev/null +++ b/src/pipecat/observers/stt_llm_tts_trace_observer.py @@ -0,0 +1,155 @@ +# +# Copyright (c) 2024–2025, Daily +# +# SPDX-License-Identifier: BSD 2-Clause License +# + +"""LLM logging observer for Pipecat.""" + +from loguru import logger + +from pipecat.frames.frames import ( + BotStartedSpeakingFrame, + BotStoppedSpeakingFrame, + LLMFullResponseEndFrame, + LLMFullResponseStartFrame, + TranscriptionFrame, + TTSStartedFrame, + TTSStoppedFrame, + UserStartedSpeakingFrame, + UserStoppedSpeakingFrame, +) +from pipecat.observers.base_observer import BaseObserver, FramePushed +from pipecat.processors.frame_processor import FrameDirection +from pipecat.services.llm_service import LLMService +from pipecat.services.stt_service import STTService +from pipecat.services.tts_service import TTSService +from pipecat.transports.base_output import BaseOutputTransport + + +class STTLLMTTSTraceObserver(BaseObserver): + """Observer to basic STT, LLM, & TTS activity to the console. + + Logs all frame instances of: + + - UserStartedSpeakingFrame + - TranscriptionFrame + + - LLMFullResponseStartFrame + - LLMFullResponseEndFrame + + - TTSStartedFrame + - TTSStoppedFrame + - BotStartedSpeakingFrame + - BotStoppedSpeakingFrame + """ + + def __init__(self): + """Initialize frame start times to calculate span times.""" + super().__init__() + self._last_user_started_speaking_frame_time = 0 + self._last_transcription_frame_time = 0 + + self._last_llm_response_start_frame_time = 0 + + self._last_tts_started_frame_time = 0 + self._last_tts_stopped_frame_time = 0 + self._last_bot_started_speaking_frame_time = 0 + + self._arrow = "→" + + async def on_push_frame(self, data: FramePushed): + """Handle frame push events and log STT, LLM, & TTS activities. + + Args: + data: The frame push event data containing source, destination, + frame, direction, and timestamp information. + """ + src = data.source + dst = data.destination + frame = data.frame + direction = data.direction + timestamp = data.timestamp + + time_sec = timestamp / 1_000_000_000 + + if isinstance(src, BaseOutputTransport): + # Trace STT + if isinstance(frame, UserStartedSpeakingFrame): + self._handle_UserStartedSpeakingFrame(src, dst, frame, time_sec) + elif isinstance(frame, UserStoppedSpeakingFrame): + self._handle_UserStoppedSpeakingFrame(src, dst, frame, time_sec) + # TTS + if isinstance(dst, TTSService): + if isinstance(frame, BotStartedSpeakingFrame): + self._handle_BotStartedSpeakingFrame(src, dst, frame, time_sec) + elif isinstance(frame, BotStoppedSpeakingFrame): + self._handle_BotStoppedSpeakingFrame(src, dst, frame, time_sec) + + # STT + elif isinstance(src, STTService): + if isinstance(frame, TranscriptionFrame): + self._handle_TranscriptionFrame(src, dst, frame, time_sec) + + # Trace LLM + elif isinstance(src, LLMService): + if isinstance(frame, LLMFullResponseStartFrame): + self._handle_LLMFullResponseStartFrame(src, dst, frame, time_sec) + elif isinstance(frame, LLMFullResponseEndFrame): + self._handle_LLMFullResponseEndFrame(src, dst, frame, time_sec) + + # Trace TTS + elif isinstance(src, TTSService): + if isinstance(frame, TTSStartedFrame): + self._handle_TTSStartedFrame(src, dst, frame, time_sec) + elif isinstance(frame, TTSStoppedFrame): + self._handle_TTSStoppedFrame(src, dst, frame, time_sec) + + # STT frame handlers + def _handle_UserStartedSpeakingFrame(self, src, dst, frame, time_sec): + self._last_user_started_speaking_frame_time = time_sec + logger.debug(f"🙂🟢 UserStartedSpeakingFrame: {src} {self._arrow} {dst} at {time_sec:.2f}s") + + def _handle_UserStoppedSpeakingFrame(self, src, dst, frame, time_sec): + logger.debug(f"🙂🔴 UserStoppedSpeakingFrame: {src} {self._arrow} {dst} at {time_sec:.2f}s") + + def _handle_TranscriptionFrame(self, src, dst, frame, time_sec): + self._last_transcription_frame_time = time_sec + logger.debug(f"🙂📝 TranscriptionFrame: {src} {self._arrow} {dst} at {time_sec:.2f}s") + + if 0 != self._last_user_started_speaking_frame_time: + stt_generation_time = time_sec - self._last_user_started_speaking_frame_time + self._last_user_started_speaking_frame_time = 0 + logger.info(f"📝⏰ STT span: {stt_generation_time:.4f}s") + + # LLM frame handlers + def _handle_LLMFullResponseStartFrame(self, src, dst, frame, time_sec): + self._last_llm_response_start_frame_time = time_sec + logger.debug( + f"🧠🟢 LLMFullResponseStartFrame: {src} {self._arrow} {dst} at {time_sec:.2f}s" + ) + + def _handle_LLMFullResponseEndFrame(self, src, dst, frame, time_sec): + logger.debug(f"🧠🔴 LLMFullResponseEndFrame: {src} {self._arrow} {dst} at {time_sec:.2f}s") + llm_time = time_sec - self._last_llm_response_start_frame_time + logger.info(f"🧠⏰ LLM span: {llm_time:.4f}s") + + # TTS frame handlers + def _handle_TTSStartedFrame(self, src, dst, frame, time_sec): + self._last_tts_started_frame_time = time_sec + logger.debug(f"📢🟢 TTSStartedFrame: {src} {self._arrow} {dst} at {time_sec:.2f}s") + + def _handle_TTSStoppedFrame(self, src, dst, frame, time_sec): + self._last_tts_stopped_frame_time = time_sec + tts_time = time_sec - self._last_tts_started_frame_time + logger.debug(f"📢🔴 TTSStoppedFrame: {src} {self._arrow} {dst} at {time_sec:.2f}s") + logger.info(f"📢⏰ TTS generation span: {tts_time:.4f}s") + + def _handle_BotStartedSpeakingFrame(self, src, dst, frame, time_sec): + self._last_bot_started_speaking_frame_time = time_sec + tts_time = time_sec - self._last_tts_started_frame_time + logger.debug(f"🤖🟢 BotStartedSpeakingFrame: {src} {self._arrow} {dst} at {time_sec:.2f}s") + logger.info(f"📢⏰ TTS to first speech span: {tts_time:.4f}s") + + def _handle_BotStoppedSpeakingFrame(self, src, dst, frame, time_sec): + logger.debug(f"🤖🔴 BotStoppedSpeakingFrame: {src} {self._arrow} {dst} at {time_sec:.2f}s")