Nim interface to WEBRTC Voice Activity Detector (VAD).

Actual code for VAD is developed by the Google for the WebRTC project and is one the best freely available VAD engine.

Installation

• $ nimble install https://gitlab.com/eagledot/nim-webrtcvad

  OR

• $ nimble install webrtcvad

Usage:

1.Create and Initialize a VAD object

import webrtcvad
var vad: vadObj
var code = initVad(vad)

2.Set its aggressiveness mode, which is an integer between 0 and 3. 0 is the least aggressive about filtering out non-speech, 3 is the most aggressive.

discard setMode(vad,3'i32)  #can use return value to check if mode set successfully

3.Test it on a short segment of audio. The WebRTC VAD only accepts 16-bit mono PCM audio, sampled at 8000, 16000, 32000 or 48000 Hz. A frame must be either 10, 20, or 30 ms in duration:

let sampleRate = 16000*2
let frameDuration = 30 #in ms
#for this fakeData should return 0 
let fakeData = newSeq[int16](int((frameDuration*sampleRate)/1000))
#returns 1 if voice activity detected else 0 but returns -1 when some error occured.
code = isSpeech(vad,fakeData,sampleRate)