A dead simple STT library in Node
Sonus lets you quickly and easily add a VUI (Voice User Interface) to any hardware or software project. Just like Alexa, Google Assistant, and Siri, Sonus is always listening offline for a customizable hotword. Once that hotword is detected your speech is streamed to the cloud recognition service of your choice - then you get the results in realtime.
- Linux - most major distros (Including Raspbian)
- macOS
- Windows
- Google Cloud Speech
- Alexa Voice Services
- Wit.ai
- Microsoft Cognitive Services
- Houndify
npm install --save sonus
Generally, running npm install should suffice. This module however, requires you to install SoX.
Recommended: use arecord, which comes with most linux distros.
Alternatively:
sudo apt-get install sox libsox-fmt-all
brew install sox
Configure out cloud speech recognition system of choice, like Google Cloud Speech API.
Note: You need to use the GOOGLE_APPLICATION_CREDENTIALS environment variable for your JSON keyfile, or check the examples to see how you can pass in the keyflie path.
Add sonus and said recognizer:
const Sonus = require('sonus')
const speech = require('@google-cloud/speech')
const client = new speech.SpeechClient()Add your keyword and initialize Sonus with a Snowboy hotword:
const hotwords = [{ file: 'resources/snowboy.umdl', hotword: 'snowboy' }]
const sonus = Sonus.init({ hotwords }, client)Create your own Alexa in less than a tweet:
Sonus.start(sonus)
sonus.on('hotword', (index, keyword) => console.log("!"))
sonus.on('final-result', console.log)This project uses semantic versioning as of v0.1.0
Follow these instructions.
Sonus uses Snowboy for offline hotword recognition. You can use their website or API to train a model for a new hotword. Hotword training must occur online through their web service.
Built #withsonus
If you've build a project with Sonus send a PR and include it here!
Evan Cohen: @_evnc
Ashish Chandwani: @ashishschandwa1
Licensed under MIT.
