Package Exports

electron-vosk-speech

This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (electron-vosk-speech) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

Readme

Electron-vosk-speech

Lightweight Speech Recognition Library for Electron. Based on nodejs-speech-kiosk-usercase and vosk-api.

Requirements: Python 3.8

Install: yarn add electron-vosk-speech To install python, read this guide. pip3 install vosk
Then, you need to download vosk's speech model: For example:

#download and save model to models/ru-small
cd node_modules/electron-vosk-speech/src/vosk
sh _model_loader.sh https://alphacephei.com/vosk/models/vosk-model-small-ru-0.4.zip ru

Or:

# download and save big russian model to models/ru-Ru
# then "ru" are used as languageCode in the options of Recognizer class
# so, you can download any models, and use it
cd node_modules/electron-vosk-speech/src/vosk
sh _model_loader.sh https://alphacephei.com/vosk/models/vosk-model-ru-0.10.zip ru

You can find list of vosk pretrained models here.

Example usage:

Add this to Electron's render process:

const {Recognizer} = require('electron-vosk-speech')
const {ipcRenderer} = require('electron')

const Rec = new Recognizer({
    ipcRenderer, 
    onSpeechRecognized: res => console.log('РЕЗУЛЬТАТ! ' + JSON.stringify(res)), 
    onSpeechStart: () => console.log('ГОВОРИТ!'), // fires on speech started
    onSpeechEnd: () => console.log('ЗАМОЛЧАЛ!'), // fires on speech ended
    options: {
        isSpeech2Text = true,
        autoInit = true,
        forced = true, // forced start recording
        idleDelay = 5000,
        languageCode = 'ru',
        harkOptions = {},
        gsFormat = false // if true, returns result in GoogleSpeech format
        // for backward compatibility with solutions based on GoogleSpeech
    }
})

// Rec.startAll() - start listening, recording and recognize
// Rec.stopAll() - stop listening, recording and recognize

// Rec.startListening() - start listening
// Rec.stopListening() - stop listening

// Rec.stopRecognize() - stop recording and recognize
// Rec.startRecognize() - start recording and recognize

Add this to Electron's main process:

const { app, BrowserWindow, ipcMain } = require('electron')
const {speechSaverHandler} = require('electron-vosk-speech/src/utils')

const win = new BrowserWindow({
    width: 800,
    height: 600,
    webPreferences: {
      nodeIntegration: true
    }
})

// than add this handler to your 'will-download' event
const projectPath = app.getAppPath()
win.webContents.session.on('will-download', function(...rest){
    speechSaverHandler(projectPath, ...rest)
})

Enjoy!