I'd like to get the start and end times of each utterance an agent and a user makes. What would be the best way to do that?

Thanks for reaching out, Abdullah. I'm an AI assistant and I'm here to help. Give me a moment while I research this topic.

from a live session

I'm happy to help you get the start and end times of each utterance during a live session! Here are some friendly pointers to guide you through: For Agent Utterances (AI Speech): • Make sure to enable

use_tts_aligned_transcript=True

in your

AgentSession

configuration. • Then, implement the

transcription_node

method in your agent class to handle

TimedString

objects, which include helpful

start_time

and

end_time

fields.

async def transcription_node(
    self, text: AsyncIterable[str | TimedString], model_settings: ModelSettings
) -> AsyncGenerator[str | TimedString, None]:
    async for chunk in text:
        if isinstance(chunk, TimedString):
            <http://logger.info|logger.info>(f"TimedString: '{chunk}' ({chunk.start_time} - {chunk.end_time})")
        yield chunk

Just a quick note: this feature is experimental and currently available only in Python. For User Utterances: • You can tap into the

user_input_transcribed

event to get transcription data with timing info, provided your STT provider supports it (like the Gladia STT plugin). • Also, telemetry attributes such as

lk.start_time

and

lk.end_time

are great for tracking speaking spans. Frontend Access: • Try using

useVoiceAssistant()

for agent transcriptions and

useTrackTranscription()

for user transcriptions. • By combining these and sorting them by

firstReceivedTime

, you can beautifully reconstruct the conversation timeline. If you'd like, I'm here to walk you through any part of the implementation or provide more details. Just let me know! Sources: Text and transcriptions | LiveKit Docs | livekit-plugins/livekit-plugins-google/livekit/plugins/google/stt.py | complex-agents/nutrition-assistant/nutrition-assistant-frontend/hooks/useCo... | Understanding Transcript Timing in Voice Conversations