snakers4/silero-vad

Install

npm install @wfloat/wfloat-web

Voice activity detection

Serve your recording at /recording.wav.

import { loadVoiceActivityDetection } from '@wfloat/wfloat-web';

const model = await loadVoiceActivityDetection('snakers4/silero-vad');
const audio = await fetch('/recording.wav').then(response => response.blob());
const result = await model.detect(audio).result();
console.log(result.segments);
await model.unload();

Voice activity detection reference · Downloads and cleanup

Install

npm install @wfloat/react-native-wfloat
cd ios && pod install

Rebuild the app after installation. Native setup.

Voice activity detection

Replace the URI with a local recording accessible to your app.

import { loadVoiceActivityDetection } from '@wfloat/react-native-wfloat';

const model = await loadVoiceActivityDetection('snakers4/silero-vad');
const audio = { uri: 'file:///path/to/recording.wav' };
const result = await model.detect(audio).result();
console.log(result.segments);
await model.unload();

Voice activity detection reference · Downloads and cleanup

Install

pip install wfloat

Voice activity detection

Use a PCM WAV recording.

from wfloat import load_voice_activity_detection

with load_voice_activity_detection("snakers4/silero-vad") as model:
    result = model.detect("recording.wav")
    for segment in result.segments:
        print(segment.start_ms, segment.end_ms)

Voice activity detection reference · Downloads and cleanup