A speech-to-text library for React Native.
yarn add @react-native-voice/voice
# or
npm i @react-native-voice/voice --save
Link the iOS package
npx pod-install
Manually or automatically link the NativeModule
react-native link @react-native-voice/voice
- In
android/setting.gradle
...
include ':@react-native-voice_voice', ':app'
project(':@react-native-voice_voice').projectDir = new File(rootProject.projectDir, '../node_modules/@react-native-voice/voice/android')
- In
android/app/build.gradle
...
dependencies {
...
compile project(':@react-native-voice_voice')
}
- In
MainApplication.java
import android.app.Application;
import com.facebook.react.ReactApplication;
import com.facebook.react.ReactPackage;
...
import com.wenkesj.voice.VoicePackage; // <------ Add this!
...
public class MainActivity extends Activity implements ReactApplication {
...
@Override
protected List<ReactPackage> getPackages() {
return Arrays.<ReactPackage>asList(
new MainReactPackage(),
new VoicePackage() // <------ Add this!
);
}
}
-
Drag the Voice.xcodeproj from the @react-native-voice/voice/ios folder to the Libraries group on Xcode in your poject. Manual linking
-
Click on your main project file (the one that represents the .xcodeproj) select Build Phases and drag the static library, lib.Voice.a, from the Libraries/Voice.xcodeproj/Products folder to Link Binary With Libraries
Full example for Android and iOS.
import Voice from '@react-native-voice/voice';
import React, {Component} from 'react';
class VoiceTest extends Component {
constructor(props) {
Voice.onSpeechStart = this.onSpeechStartHandler.bind(this);
Voice.onSpeechEnd = this.onSpeechEndHandler.bind(this);
Voice.onSpeechResults = this.onSpeechResultsHandler.bind(this);
}
onStartButtonPress(e){
Voice.start('en-US');
}
...
}
Static access to the Voice API.
All methods now return a new Promise
for async/await
compatibility.
Method Name | Description | Platform |
---|---|---|
Voice.isAvailable() | Checks whether a speech recognition service is available on the system. | Android, iOS |
Voice.start(locale) | Starts listening for speech for a specific locale. Returns null if no error occurs. | Android, iOS |
Voice.stop() | Stops listening for speech. Returns null if no error occurs. | Android, iOS |
Voice.cancel() | Cancels the speech recognition. Returns null if no error occurs. | Android, iOS |
Voice.destroy() | Destroys the current SpeechRecognizer instance. Returns null if no error occurs. | Android, iOS |
Voice.removeAllListeners() | Cleans/nullifies overridden Voice static methods. |
Android, iOS |
Voice.isRecognizing() | Return if the SpeechRecognizer is recognizing. | Android, iOS |
Voice.getSpeechRecognitionServices() | Returns a list of the speech recognition engines available on the device. (Example: ['com.google.android.googlequicksearchbox'] if Google is the only one available.) |
Android |
Callbacks that are invoked when a native event emitted.
Event Name | Description | Event | Platform |
---|---|---|---|
Voice.onSpeechStart(event) | Invoked when .start() is called without error. |
{ error: false } |
Android, iOS |
Voice.onSpeechRecognized(event) | Invoked when speech is recognized. | { error: false } |
Android, iOS |
Voice.onSpeechEnd(event) | Invoked when SpeechRecognizer stops recognition. | { error: false } |
Android, iOS |
Voice.onSpeechError(event) | Invoked when an error occurs. | { error: Description of error as string } |
Android, iOS |
Voice.onSpeechResults(event) | Invoked when SpeechRecognizer is finished recognizing. | { value: [..., 'Speech recognized'] } |
Android, iOS |
Voice.onSpeechPartialResults(event) | Invoked when any results are computed. | { value: [..., 'Partial speech recognized'] } |
Android, iOS |
Voice.onSpeechVolumeChanged(event) | Invoked when pitch that is recognized changed. | { value: pitch in dB } |
Android |
Arguably the most important part.
While the included VoiceTest
app works without explicit permissions checks and requests, it may be necessary to add a permission request for RECORD_AUDIO
for some configurations.
Since Android M (6.0), user need to grant permission at runtime (and not during app installation).
By default, calling the startSpeech
method will invoke RECORD AUDIO
permission popup to the user. This can be disabled by passing REQUEST_PERMISSIONS_AUTO: true
in the options argument.
If you're running an ejected expo/expokit app, you may run into issues with permissions on Android and get the following error host.exp.exponent.MainActivity cannot be cast to com.facebook.react.ReactActivity startSpeech
. This can be resolved by prompting for permssion using the expo-permission
package before starting recognition.
import { Permissions } from "expo";
async componentDidMount() {
const { status, expires, permissions } = await Permissions.askAsync(
Permissions.AUDIO_RECORDING
);
if (status !== "granted") {
//Permissions not granted. Don't show the start recording button because it will cause problems if it's pressed.
this.setState({showRecordButton: false});
} else {
this.setState({showRecordButton: true});
}
}
Notes on Android
Even after all the permissions are correct in Android, there is one last thing to make sure this libray is working fine on Android. Please make sure the device has Google Speech Recognizing Engine such as com.google.android.googlequicksearchbox
by calling Voice.getSpeechRecognitionServices()
. Since Android phones can be configured with so many options, even if a device has googlequicksearchbox engine, it could be configured to use other services. You can check which serivce is used for Voice Assistive App in following steps for most Android phones:
Settings > App Management > Default App > Assistive App and Voice Input > Assistive App
Above flow can vary depending on the Android models and manufactures. For Huawei phones, there might be a chance that the device cannot install Google Services.
How can I get com.google.android.googlequicksearchbox
in the device?
Please ask users to install Google Search App.
Need to include permissions for NSMicrophoneUsageDescription
and NSSpeechRecognitionUsageDescription
inside Info.plist for iOS. See the included VoiceTest
for how to handle these cases.
<dict>
...
<key>NSMicrophoneUsageDescription</key>
<string>Description of why you require the use of the microphone</string>
<key>NSSpeechRecognitionUsageDescription</key>
<string>Description of why you require the use of the speech recognition</string>
...
</dict>
Please see the documentation provided by ReactNative for this: PermissionsAndroid
- @asafron
- @BrendanFDMoore
- @brudny
- @chitezh
- @ifsnow
- @jamsch
- @misino
- @Noitidart
- @ohtangza & @hayanmind
- @rudiedev6
- @tdonia
- @wenkesj