To switch

RoomOutputOptions

audio from enabled (

true

) to disabled (

false

) while the session is running, you can use the

session.output.set_audio_enabled()

method dynamically during runtime. Here is an example:

# Start with audio enabled
session = AgentSession(...)
await session.start(
    room=ctx.room,
    room_output_options=RoomOutputOptions(audio_enabled=True),
    # ... other options
)

# Later, disable audio output while session is running
session.output.set_audio_enabled(False)

# Re-enable audio output if needed
session.output.set_audio_enabled(True)

If you need to control this remotely, you can implement RPC methods. For example:

@ctx.room.local_participant.register_rpc_method("toggle_output")
async def on_toggle_output(data: rtc.RpcInvocationData) -> None:
    if data.payload == "audio_on":
        session.output.set_audio_enabled(True)
    elif data.payload == "audio_off":
        session.output.set_audio_enabled(False)

This approach allows you to dynamically control the audio output during the session's lifetime. Sources: Text and transcriptions | LiveKit Docs | livekit-agents/livekit/agents/voice/io.py