Barge-in

Talk over the avatar and it stops mid-sentence, then listens: that is barge-in.

• The CLI and the Python agent: the voice model’s turn detection hears you start talking and cancels its reply; LiveKit Agents then clears the avatar’s buffered audio and frames, so the mouth stops with the voice. In the Python agent it is interrupt_response=True on the turn detection, shown in Voice agent in Python.
• Your own loop in Python: when your speech detection fires, call interrupt() on the runtime and clear any reply audio you still hold. run() carries on with idle frames. With OpenAI Realtime, the event to watch is input_audio_buffer.speech_started:

# excerpt: barge-in, in the event loop of conversation.py
elif event.type == "input_audio_buffer.speech_started":
    # the user started talking: drop the rest of the reply
    runtime.interrupt()

• On the device: interrupt() in Swift and Flutter; resetState(true) for Expression 2 and resetAudio() for Essence 2 on Android (Companion app).
• Pipecat: BitHumanVideoService handles InterruptionFrame: the avatar drops the reply in flight and goes back to idle (Pipecat).
• A cloud avatar without the plugin: perform the RPC lk.clear_buffer on the avatar (Cloud avatar).

Barge-in does not change billing: a session bills active session time, talking or idle, to the second.