react-native-android-speech-recognizer

The purpose of this module is to provide access to Android's SpeechRecognizer API for React Native apps.

This module isn't meant to abstract the API. Higher level modules could be written to do that.

The SpeechRecognizer can be used to integrate voice recognition into your app rather than using the default UI.

Getting started

$ npm install react-native-android-speech-recognizer --save

Mostly automatic installation

$ react-native link react-native-android-speech-recognizer

Manual installation

Android

1. Open up android/app/src/main/java/[...]/MainActivity.java

• Add import de.siteof.rn.androidspeechrecognizer.RNAndroidSpeechRecognizerPackage; to the imports at the top of the file
• Add new RNAndroidSpeechRecognizerPackage() to the list returned by the getPackages() method

2. Append the following lines to android/settings.gradle:

include ':react-native-android-speech-recognizer'
project(':react-native-android-speech-recognizer').projectDir = new File(rootProject.projectDir, 	'../node_modules/react-native-android-speech-recognizer/android')

3. Insert the following lines inside the dependencies block in android/app/build.gradle:

  compile project(':react-native-android-speech-recognizer')

Permissions

To use this library you will need the RECORD_AUDIO permission.

Insert the following in android/app/src/AndroidManifest.xml:

<uses-permission android:name="android.permission.RECORD_AUDIO" />

Usage

The API follows Android's SpeechRecognizer and RecognizerIntent

Some changes are necessary due to the way React Native works:

• Methods can only return promises (or nothing) due to being asynchronous
• Types are restricted by the what React Native supports
• Callbacks can't be called multiple times (events can)

import {
  SpeechRecognizer,
  RecognizerIntent,
  RecognitionListener
} from 'react-native-android-speech-recognizer';

const recognise = options => new Promise(async (resolve, reject) => {
  const available = await SpeechRecognizer.isRecognitionAvailable();
  if (!available) {
    reject("not available");
  }
  const recognizer = await SpeechRecognizer.createSpeechRecognizer();
  recognizer.setRecognitionListener({
    onError: event => reject("Failed with error code: " + event.error),
    onResults: event => {
      const recognition = event.results[SpeechRecognizer.RESULTS_RECOGNITION];
      const bestRecognition = recognition[0];
      resolve(bestRecognition);
    }
  });
  recognizer.startListening(RecognizerIntent.ACTION_RECOGNIZE_SPEECH, {});
});

recognise().then(bestRecognition => {
  console.log("recognised:", resultTextToEvent(bestRecognition));
}).catch(error => {
  console.log("error:", error);
});

You could also request partial results like so:

recognizer.setRecognitionListener({
  // ...
  onPartialResults: event => {
    const recognition = event.partialResults[SpeechRecognizer.RESULTS_RECOGNITION];
    const bestRecognition = recognition[0];
    console.log("best recognition so far:", bestRecognition);
  }
});

recognizer.startListening(
  RecognizerIntent.ACTION_RECOGNIZE_SPEECH, {
    [RecognizerIntent.EXTRA_PARTIAL_RESULTS]: true
  }
);