babbler

Wrapper for the Google Chrome speech synthesis and web speech recognition APIs.

Why?

Nicer API. Just that. You can very much use the vanilla speech synthesis and speech recognition APIs if you like, this just provides a nicer way to interact with them. The final decision is up to you.

## Use it

It was developed entirely with ES2015 but compiled to ES5 using babel. You will need to import it using something like browserify, webpack, etc.

Babbler exports two functions: speak and recognizer.

It does not import any es6-promise polyfill, that is up to you.

speak

import { speak } from 'babbler';

speak("Hey, it's me!").then(e => {
  console.log('Browser finished speaking!', e);
});

speak takes the string to speak and returns a promise that will be fulfilled when the browser ends speaking. The promise will be rejected if the browser does not support speechSynthesis API.

recognizer

Instances a new speech recognizer.

import { recognizer } from 'babbler';

let r = recognizer({
  onStart(e) {
    // foo
  },

  onEnd(e) {
    // foo
  },

  onResult(e) {
    // foo
  },

  onError(e) {
    // foo
  },

  continuous: false,
  interimResults: false
});

// Start speech recognition
r.start();

// Stop speech recognition
r.stop();

// redefine onStart callback
r.onStart(e => {
  // you can redefine the event handlers
});

// All parameters are optional in the recognizer constructor.
// In fact, you can initialize a default recognizer like this:
let r2 = recognizer();

Tests

By now you must have realized the library has no tests. Why? Because everything would need to be mocked and test how a wrapper behaves with a mock does not make any sense. So don't expect any tests. It's just a wrapper, anyway.