To save the full call transcript using

session.history

in Python when the call ends or the user leaves the room, here’s a helpful approach you can try:

from datetime import datetime
import json
import logging

async def entrypoint(ctx: JobContext):
    session = AgentSession(
        stt=deepgram.STT(),
        llm=openai.LLM(),
        tts=cartesia.TTS(),
        vad=silero.VAD.load(),
    )

    # Define shutdown callback for saving transcript
    async def write_transcript():
        try:
            # Ensure any pending agent responses are flushed
            await session.aclose()

            current_date = datetime.now().strftime("%Y%m%d_%H%M%S")
            filename = f"/tmp/transcript_{ctx.room.name}_{current_date}.json"
            
            # Convert history to dictionary and save
            transcript_data = session.history.to_dict()
            with open(filename, 'w', encoding='utf-8') as f:
                json.dump(transcript_data, f, indent=2)

            <http://logging.info|logging.info>(f"Transcript saved to {filename}")
        except Exception as e:
            logging.error(f"Failed to save transcript: {e}")

    # Register the shutdown callback
    ctx.add_shutdown_callback(write_transcript)

    # Start the session
    await session.start(agent=MyAgent(), room=ctx.room)

Key Points: • Make sure to use

session.aclose()

so all conversation items are properly finalized before saving. • It’s a good idea to wrap file operations in a

try-except

block to gracefully handle any unexpected errors. • For production, consider saving the transcript to a persistent and easily accessible location, like cloud storage. If you have any questions or would like me to walk you through this further, just let me know—I’m here to help! Sources: basics/exit_message.py | Inside a session | LiveKit Docs | livekit-plugins/livekit-plugins-openai/livekit/plugins/openai/stt.py | Session recording and transcripts | LiveKit Docs | Workflows | LiveKit Docs