@audio/vad

Voice activity detection — frame-level speech/non-speech decision, plus the speech-presence primitives that drive spectral denoise.

Classical, deterministic, no model weights. vad() gives a per-frame active/inactive track; spp() and ddSnr() are the per-bin estimators that OM-LSA / Wiener / IMCRA gain rules consume.

import { vad, spp, ddSnr } from '@audio/vad'

let { active, times, hop, frameSize } = vad(signal, { fs: 48000 })
// active: Uint8Array — 1 where speech is present
// times:  Float32Array — frame-start time (s) for each flag

vad(data, opts?)

Per-frame decision: a frame is active iff its energy sits snrTh dB above the noise floor and its spectral flatness is below flatTh (speech is tonal → low flatness; noise is flat → high). The floor is the 10th-percentile frame energy plus bias dB — robust on clips where any short window may be entirely speech, avoiding the "speech eats its own floor" failure of exponential smoothing.

| opt | default | meaning | |---|---|---| | fs | 44100 | sample rate (Hz) | | frameSize | 1024 | STFT frame length | | hopSize | frameSize/2 | hop between frames | | snrTh | 6 | dB above floor to count as active | | flatTh | 0.4 | spectral-flatness ceiling for "tonal" | | bias | 5 | dB added to the percentile floor |

Returns { active: Uint8Array, times: Float32Array, hop, frameSize }.

spp(mag, noisePsd, opts?)

Per-bin speech-presence probability from a-priori SNR ξ: p = ξ / (1 + ξ) (Gaussian model, q-prior 0.5). Bind to a noise PSD (e.g. minStats/imcra from @audio/denoise). opts.xiMin floors ξ (default 0.0316, −15 dB).

ddSnr(mag, noisePsd, prevGain, prevMag, alpha?)

Decision-directed a-priori SNR (Ephraim & Malah 1984), recursively smoothed with alpha (default 0.98). The ξ̂ estimate feeding Wiener / MMSE / OM-LSA gains.

Notes

STFT via @audio/stft. Also re-exported from @audio/denoise for restoration pipelines. MIT.