B4A Library SpeechToText - Continuous Offline Voice Recognition

This is a wrapper of Acephei VOSK , With this, you can add continuous offline speech recognition feature to your application,

NOTE:

1. As it works offline the app should be complied with the voice model. It will increase the app size by 30-40Mb.
2. The accuracy depends on the voice model. You can train your own voice model. For more details check the models download link below.
3. Remember to add RECORD_AUDIO permission.

How to use:

1. Download the required voice model from here.
2. Change the file name to a simple one like "model.zip"
3. Copy it to the Files folder of your project.
4. Now to use that model check the attached example.

SpeechToText

Author: @Biswajit
Version: 1.5

• SpeechToText
  • Events:
    • Error (message As String)
    • FinalResult (text As String)
    • MicrophoneBuffer (buffer() As Byte)
    • PartialResult (text As String)
    • Paused (paused As Boolean)
    • ReadyToListen
    • ReadyToListenEx new
    • ReadyToRead
    • Restarted
    • Result (text As String)
  • Fields:
    • sampleRate As Int
      Default 16000
  • Functions:
    • cancel As Boolean
      Cancel microphone recognition. Do not post any new events, simply cancel processing.
      Does nothing if recognition is not active.
      Return type: @return:true if recognition was actually stopped
    • FeedExternalBuffer (ExBuffer As Byte()) new
      For recognizing the external audio buffer, feed the buffer here.
      ExBuffer: The external audio byte buffer.
    • Initialize (eventName As String, modelPath As String)
      Initialize the object.
      eventName: The event name prefix.
      modelPath: The model folder path.
    • pause (pause As Boolean)
      Pause microphone recognition.
      pause: Pass true to pause and false to continue.
    • prepareAudioFile (audioPath As String, predefinedWords As String)
      Prepare the audio file for recognition. On success Eventname_ReadyToRead event will be raised.
      Call startReading to start reading the file.
      audioPath: Audio file path.
      predefinedWords: Add some predefined words/phrase as JSON string. Can be blank.
    • prepareListenerEx (predefinedWords As String) new
      Prepare the listener for external audio buffer. On success Eventname_ReadyToListenEx event will be raised.
      Call startListeningEx to start listening.
      predefinedWords: Add some predefined words/phrase as JSON string. Can be blank.
    • prepareMicrophone (predefinedWords As String)
      Prepare the microphone for listening. On success Eventname_ReadyToListen event will be raised.
      Call startListening to start listening.
      predefinedWords: Add some predefined words/phrase as JSON string. Can be blank.
    • reset
      Resets microphone recognizer in a thread, starts microphone recognition over again
    • shutdown
      Shutdown the microphone recognizer and release the recorder.
      Call this on activity or service closing event.
    • startListening (timeout As Int) As Boolean
      Starts microphone recognition. After specified timeout listening stops and the
      endOfSpeech signals about that. Does nothing if recognition is active.
      timeout: timeout in milliseconds to listen. -1 = infinite;
      Return type: @return:true if recognition was actually started
    • startListeningEx As Boolean new
      Starts external audio buffer recognition.
      Return type: @return:true if recognition was actually started
    • startReading (timeout As Int) As Boolean
      Starts file recognition. After specified timeout listening stops and the
      endOfSpeech signals about that. Does nothing if recognition is active.
      timeout: timeout in milliseconds to listen. -1 = infinite;
      Return type: @return:true if recognition was actually started
    • stop As Boolean
      Stops microphone/file recognition. Listener should receive final result if there is
      any. Does nothing if recognition is not active.
      Call this on activity or service closing event.
      Return type: @return:true if recognition was actually stopped

Downloads:

1. Library
2. Example
3. Voice Model
4. Test app

Update:

• Version 1.1:
  1. Added audio file to text functionality. (For now only WAV format is supported)
  2. Added predefined word/phrase detection functionality.
  3. Merged startListening and startListening2 together. Pass -1 for continuous recognition.
• Version 1.2:
  1. Added MicrophoneBuffer event where you will receive the microphone audio buffer while using voice recognition.
• Version 1.3:
  1. Added method to change the sampling rate.
• Version 1.4:
  1. Fixed the app crashing issue while calling shutdown without stating the recognizer
• Version 1.5:
  1. Added option to feed external audio buffer. Instead of using the internal audio recorder you can feed external audio buffer from another audio source.
     (Check the latest example project)
  2. Updated VOSK and JNA library. (Please delete old dependencies before coping the new ones.)

If you like my work, please donate. Your donations will encourage me to add more features in the future.

 

[Khalid.]

You have to add your own logic to add space when the speaker is silent.

Using (Voice Activity Detection), specify when it is off. Space append to text
https://en.m.wikipedia.org/wiki/Voice_activity_detection

And sometimes there is an extended word like:

((oooooooooh))) my god

how is the letter repeated?

 

[Biswajit]

Using (Voice Activity Detection), specify when it is off. Space append to text
https://en.m.wikipedia.org/wiki/Voice_activity_detection

And sometimes there is an extended word like:

how is the letter repeated?

I did not understand your query. It's better to post your question on the forum directly so that the other members can also check.

 

[Khalid.]

I did not understand your query

 

[Khalid.]

I did not understand your query

Perhaps this video shows what I mean:
detection if the sound is dashed when it is voice detected ( i ) is set and when voice not detected ( _ ) is set. In your library I need a voice not detected to append ((space)) with the converted text
yes . I already posted
https://www.b4x.com/android/forum/threads/synchronize-text-and-audio-together.143692/
@stevel05

 

[Biswajit]

Why are you sharing these videos? If you want to add the functionality you have to think of a solution. If you are facing a problem post that to the forum directly. I am not gonna check the code from the video and do it for you as it's not related to this library.

This is a simple example of adding space/dot when there is no speech detected by the library,

Sub STT_PartialResult(text As String)
    If text.Trim.length > 0 Then
        partialResultBox.Text = text
        timer.Enabled = False
    Else
        If StopBtn.Enabled Then timer.Enabled = True
    End If
End Sub

Sub timer_Tick
    ''add space/dot when the partial result length is 0
    resultBox.Text = resultBox.Text & " . "
End Sub

 

[Khalid.]

This is a simple example of adding space/dot when there is no speech detected by the library,

Yes it worked. There are some continuous letter like

ooooh my god

how can the letter be extended

 

[Biswajit]

how can the letter be extended

Please create a new thread explaining your requirements.

 

[Ricks Film Restoration]

The Test-app APK file works very well! This is really great. However, I keep getting this error when implementing this speech recognizer in my app: java.lang.UnsatisfiedLinkError: Native library (com/sun/jna/android-aarch64/libjnidispatch.so) Not found in resource path (.) I have found both .so files and have tried to include them in dozens of ways, including adding the entire "lib" folder from the APK file, modifying the path in jna.jar and using "#AdditionalRes: ..\lib", but nothing works. I have also read all the threads about inclusing .so files, but still I am stuck. Please help!

 

[DonManfred]

Native library (com/sun/jna/android-aarch64/libjnidispatch.so) Not found in resource path (.)

looks like there is no 64bit libjnidispatch.so inside the library.

I have found both .so files and have tried to include them in dozens of ways

they needs to be added to the library jar file into the right subfolder. Remember that the jar is a zip-file

 

[Ricks Film Restoration]

looks like there is no 64bit libjnidispatch.so inside the library.

they needs to be added to the library jar file into the right subfolder. Remember that the jar is a zip-file

Thanks. I tried to modify the jna.jar file (with zip) by adding the folder: android-aarch64 to: com/sun/jna/ and inserted both .so files there that I derived from the lib folder within the APK "Test App" file, but no success. PS: I also added this entire lib folder to my B4A project. The APK file works fine!
What I also find strange is that there is no reference anywhere to this jna.jar (nor to the vosk-adroid.aar) file. I guess it must be somewhere hidden in the SpeechToText.jar /xml library (?)

 

[DonManfred]

library (?)

Post a downloadlink for the .so files
I´ll check

 

[DonManfred]

The needed .so is inside the AAR.
At least for
- arm64-v8a
- armeabi-v7a
- x86
- x86_64

What Device do you have?

 

[Ricks Film Restoration]

The needed .so is inside the AAR.
At least for
- arm64-v8a
- armeabi-v7a
- x86
- x86_64

What Device do you have?

Samsung Galaxy TAB S7 with Android-13. Shouldn't the AAR be added to my B4A project (or is it maybe used within the SpeechToText.jar /xml library)? Also why am I getting the afore mentioned error: java.lang.UnsatisfiedLinkError: Native library (com/sun/jna/android-aarch64/libjnidispatch.so) Not found in resource path (.) which seems to ignore the AAR?

 

[Ricks Film Restoration]

Post a downloadlink for the .so files (zipped)
I´ll check

Attached is the .so file

 

[DonManfred]

Attached is the .so file

i prefer a downloadlink where i can find the .so in it´s correct path...

 

[Ricks Film Restoration]

The .so file is taken from the APK file. Please just unzip the APK file and look in the "lib" folder:

 

[Ricks Film Restoration]

PS, You may want to try the guys' provided B4A sample project to easily replicate my problem

 

[drgottjr]

i've checked my downloads:
SpeechToText_v1.3.zip and SpeechToText_v1.4.zip (the latest, as far as i am aware) both contain all the files you need to build an app. unzip the jar and put all the files in your additional libraries folder. when you build the project, just select SpeechToText from the libraries tab in the IDE.
i've had no trouble building. see https://www.b4x.com/android/forum/threads/talk-to-the-hand-the-final-frontier.143639/#content
for a comparison between android's speech recognition capabilities and vosk's. the vosk part was built with the library right out of the box.

 

[Ricks Film Restoration]

Thank you very much. Obviously I did follow all the instructions by Biswajit. The only explanation I have that it doesn't work for me is that I am still using an old B4A version (8.3). I have not done much with B4a for ages (lost interest). Sad to read that Vosk is not as good as Google. I mainly want to use Vosk for continuous speech recognition and/or to use it as a hotword detector. I was using Snowboy for that, but it doesn't work on recent Android versions.

 

[DonManfred]

Please just unzip the APK file and look in the "lib" folder

I already told which ones are available. See #72
You are not showing any other .so here.
Sorry, i can not further help here.