You're asking about

@session.on("agent_false_interruption")

. Here's what it does and how to use it. What it means • This event fires when the system detected an interruption (agent was cut off) but no actual user speech was received within the configured timeout—i.e., a false positive from voice activity detection (VAD). • The default timeout is

false_interruption_timeout

= 2 seconds. If no final transcription arrives in that window, the interruption is considered false. What happens when it triggers • By default, the agent can resume speaking from where it left off if

resume_false_interruption

is

True

. • You can use the event to log false interruptions, adjust UI state, update metrics, or tweak detection sensitivity. Example (Python)

@session.on("agent_false_interruption")
def on_false_interruption():
    print("False interruption detected; resuming playback.")
    # custom logic: resume agent TTS, increment metric, etc.

When to handle it • If background noise or brief non-speech sounds cause unwanted truncation of agent speech. • If you need custom handling (e.g., show a UI hint, retry TTS, or change VAD parameters). If you want, tell me which SDK/language you're using and whether you need sample code to resume playback or adjust the timeout/behavior. Sources: Turns overview | LiveKit docs