Module voice_activity 

Available on non-target_family=wasm only.

Voice-activity inference over canonical audio frames: which stretches carry a human voice rather than music, noise, or silence.

Structs

Endpointer

Finds utterances in a stream. Speech opens on a frame scored at or above SPEECH_THRESHOLD and closes once silence has lasted trailing_silence. After an End it keeps listening, so the next Start opens the next utterance. An utterance still open when input ends gets no End.

SileroVadsilero

Speechsilero

What one pass over a window found.

SpeechDetectorsilero

Scores a recording’s frames and records them in a SpeechMap.

SpeechMapsilero

Where anyone was speaking, as far as the detector has scored.

Enums

Boundary

A speech boundary, in canonical samples counted from the first Endpointer::push.

VoiceActivityError

Constants

LEAD_IN_FRAMES

Frames to keep before a Boundary::Start, so an utterance does not open mid-syllable. StopOnSilence keeps this many.

SPEECH_THRESHOLD

Traits

VoiceActivityDetector