Voice Activity Detection - RunAnywhere Documentation

RunAnywhere.vad answers “is someone talking right now”. It is what a voice interface uses to decide when a turn has ended.

const result = await RunAnywhere.vad.detect(AudioInputs.file(path))

Signatures

detect(audio: AudioInput, options?: VadOptions): Promise<VadResult>
detectStream(audio: AsyncIterable<AudioInput>, options?: VadOptions): AsyncIterable<VadEvent>

Streaming

for await (const event of RunAnywhere.vad.detectStream(microphone())) {
  if (event.type === 'speechStart') setSpeaking(true)
  if (event.type === 'speechEnd') setSpeaking(false)
}

Building the audio iterable is the same problem as in streaming STT, and the same generator works for both.

Options

const options = {
  activationThreshold: 0.5,
  minSpeechMs: 250,
  minSilenceMs: 500,
  prefixPaddingMs: 300,
}

Field	Default	Meaning
`activationThreshold`	model default	How confident before it counts as speech
`minSpeechMs`	proto default	Ignore bursts shorter than this
`minSilenceMs`	proto default	Silence before a turn is over
`prefixPaddingMs`	proto default	Audio kept from before speech started

minSilenceMs is the one to tune. Too short and the assistant interrupts someone who paused; too long and it feels sluggish. prefixPaddingMs matters because detection lags the first syllable. Without padding, “hello” reaches the transcriber as “ello”.

You may not need this

For a spoken conversation, use the voice session. It runs VAD, transcription, generation, and speech together, with turn-taking and interruption wired up.