Voice: Module stop listening to voice in Android

[subhendukundu]

I am not sure if these are the correct behaviours or not but the events work a bit different in iOS and Android.
In Android the module stop listening to voice after calling the following methods

Voice.android.js:70 onSpeechPartialResults {value: Array(1)}
Voice.android.js:70 onSpeechPartialResults {value: Array(1)}
Voice.android.js:70 onSpeechPartialResults {value: Array(1)}
Voice.android.js:70 onSpeechPartialResults {value: Array(1)}
Voice.android.js:74 onSpeechEnd {error: false}
Voice.android.js:78 onSpeechResults {value: Array(5)}

Is it possible to make the module listening without making it stop?

Tested Android versions: 9.0.5, 8.0

[rpf5573]

me too, onSpeechEnd is called automatically after voice recognition in Android.

[ducpt2]

+1 really need this feature, the time onSpeechEnd is too short.

[abhishekarora1028]

Hey Guys,

Is that a new issue introduced in the latest build? We had the Android version of our app working fine while development, but now in the beta version, it's breaking just like the @subhendukundu described above.

Please help we need to make the app live asap.

Thanks
Abhi

[abhishekarora1028]

Hi Guys,

Any update on this, we are completely stuck with Android version of our app.. could you please get this resolved asap?

Thanks
Abhi

[Jithinqw]

I just ran the example code. The app is able to recognise even if the speech results is coming as errors, and its able to run for a long time.

[teslavitas]

I just ran the example code. The app is able to recognise even if the speech results is coming as errors, and its able to run for a long time.

@Jithinqw do you mean that it continues listening after long time of silence?

[Jithinqw]

Sorry, the process is stopping. I have to call the process again !

[lfoliveir4]

@subhendukundu Did you solved your problem?

[StarryFire]

Still no update on this? I really need this fixed for android!

[evtuw]

I don't think this problem will resolve, but i tried with some config below:

try {
await Voice.start('es_US', {
EXTRA_SPEECH_INPUT_COMPLETE_SILENCE_LENGTH_MILLIS: 30000,
EXTRA_SPEECH_INPUT_MINIMUM_LENGTH_MILLIS: 30000,
EXTRA_SPEECH_INPUT_POSSIBLY_COMPLETE_SILENCE_LENGTH_MILLIS: 30000
});
} catch (exception) {
console.log(exception, 'exception');
}

It's work about 5-6 seconds before auto stop on Android.

[safaiyeh]

Does anyone have insight on how to achieve this natively?

[nikhilbhawsinka]

Did anyone solve this? For me, it stops listening right after onSpeechResults.
If i call voice start at the end of onSpeechResults, since there is a lag, part of words spoken get missed. Would be great if someone could help.
Have tried "EXTRA_SPEECH_INPUT_MINIMUM_LENGTH_MILLIS: 30000" but it did not work

[burakharun]

These options don't work on my project too. Is there any solution for this issue.

EXTRA_SPEECH_INPUT_COMPLETE_SILENCE_LENGTH_MILLIS: 30000,
EXTRA_SPEECH_INPUT_MINIMUM_LENGTH_MILLIS: 30000,
EXTRA_SPEECH_INPUT_POSSIBLY_COMPLETE_SILENCE_LENGTH_MILLIS: 30000

[evtuw]

@nikhilbhawsinka ni
@safaiyeh sa
@brkhrn brk
I resolved this with SpeechToText NativeModules (in Android), you can try this solution.

Create file SpeechToTextModule.java in android/app/src/../apptest/SpeechToTextModule.java like this:

package com.test.apptest;

import android.app.Activity;
import android.content.Intent;
import android.speech.RecognizerIntent;
import android.util.Log;
import android.widget.Toast;

import androidx.annotation.NonNull;

import com.facebook.react.bridge.ActivityEventListener;
import com.facebook.react.bridge.BaseActivityEventListener;
import com.facebook.react.bridge.Promise;
import com.facebook.react.bridge.ReactApplicationContext;
import com.facebook.react.bridge.ReactContextBaseJavaModule;
import com.facebook.react.bridge.ReactMethod;

import java.util.ArrayList;
import java.util.HashMap;
import java.util.Locale;
import java.util.Map;

public class SpeechToTextModule extends ReactContextBaseJavaModule {
    private static final String DURATION_SHORT_KEY = "SHORT";
    private static final String DURATION_LONG_KEY = "LONG";
    private final int SPEECH_REQUEST_CODE = 123;
    private Promise mPickerPromise;

    public SpeechToTextModule(ReactApplicationContext reactContext) {
        super(reactContext);
        reactContext.addActivityEventListener(mActivityEventListener);
    }

    @Override
    public String getName() {
        return "SpeechToText";  // name to  export native module
    }

    @Override
    public Map<String, Object> getConstants() {
        final Map<String, Object> constants = new HashMap<>();
        constants.put(DURATION_SHORT_KEY, Toast.LENGTH_SHORT);
        constants.put(DURATION_LONG_KEY, Toast.LENGTH_LONG);
        return constants;
    }

    private final ActivityEventListener mActivityEventListener = new BaseActivityEventListener() {

        @Override
        public void onActivityResult(Activity activity, int requestCode, int resultCode, Intent data) {

            switch (requestCode) {
                case SPEECH_REQUEST_CODE: {
                    if (resultCode == Activity.RESULT_OK && null != data) {

                        ArrayList<String> result = data
                                .getStringArrayListExtra(RecognizerIntent.EXTRA_RESULTS);
                        mPickerPromise.resolve(result.get(0));
                    }
                    break;
                }

            }
        }
    };

    @ReactMethod
    public void speak(final Promise promise) {
        Activity currentActivity = getCurrentActivity();

        if (currentActivity == null) {
            mPickerPromise.reject("Your device is not supported!");
            mPickerPromise = null;
            return;
        }

        mPickerPromise = promise;

        Intent intent = new Intent(RecognizerIntent.ACTION_RECOGNIZE_SPEECH);
        intent.putExtra(RecognizerIntent.EXTRA_LANGUAGE_MODEL,
                RecognizerIntent.LANGUAGE_MODEL_FREE_FORM);
        intent.putExtra(RecognizerIntent.EXTRA_LANGUAGE, "en_US");
        try {
            this.getReactApplicationContext().addActivityEventListener(mActivityEventListener);
            currentActivity.startActivityForResult(intent, SPEECH_REQUEST_CODE);
        } catch (Exception e) {
            mPickerPromise.reject("Your device is not supported!");
            mPickerPromise = null;
        }
    }
}

Then create file ModuleSTT.java in android/app/src/../apptest/ModuleSTT.java like this:

package com.test.apptest;

import com.facebook.react.ReactPackage;
import com.facebook.react.bridge.NativeModule;
import com.facebook.react.bridge.ReactApplicationContext;
import com.facebook.react.uimanager.ViewManager;

import java.util.ArrayList;
import java.util.Collections;
import java.util.List;

public class ModuleSTT implements ReactPackage {
    @Override
    public List<ViewManager> createViewManagers(ReactApplicationContext reactContext) {
        return Collections.emptyList();
    }

    @Override
    public List<NativeModule> createNativeModules(
            ReactApplicationContext reactContext) {
        List<NativeModule> modules = new ArrayList<>();

        modules.add(new SpeechToTextModule(reactContext));

        return modules;
    }
}

Next, import ModuleSTT in MainApplication.java like this: import com.test.apptest.ModuleSTT;
In getPackages function, add this: packages.add(new ModuleSTT()), then react-native run-android.

Next, create file SpeechToText.js in your project like this content:

import { NativeModules } from 'react-native';
module.exports = NativeModules.SpeechToText;

Code js:

import SpeechToText from './SpeechToText.js'
if (Platform.OS === 'android') {
      await SpeechToText.speak()
        .then(async response => {
          console.log(response, 'response speech');
          await this.setState({ result: response, keySearch: response });
          // do anything you want
        })
        .catch(error => {
          console.log(error);
        });
    }

Sorry about my English and hope help you resolve this.

[safaiyeh]

@anhnd11 could you create a PR with these changes

[nikhilbhawsinka]

Hi @anhnd11, I tried but the recogniser still shuts off after a couple of seconds. Can you please let me know what needs to be done here?

[nikhilbhawsinka]

I have somehow hacked a solution, not sure if its a fit all thing.
What you can do is call _startRecognizing, in loop, this will stop it from stopping, but create an irritating beeping noise.
To mute that noise, use:

AudioManager mAudioManager =(AudioManager) this.reactContext.getSystemService(Context.AUDIO_SERVICE);
mAudioManager.adjustStreamVolume(AudioManager.STREAM_NOTIFICATION, AudioManager.ADJUST_MUTE,0);