Elisha Azaria

> Any updates? I would like to provide any help if you have any issues. Sorry, I'm busy right now. The main problem with these projects is that they don't...

If you are referring to this: https://github.com/alphacep/vosk-android-demo/issues/121 then it is a bug in the Vosk library. It seems that someone found a temporary fix: https://github.com/alphacep/vosk-android-demo/pull/212/files I'll have to check the...

This isn't Speech Recognition Service, but ACTION_RECOGNIZE_SPEECH. Sayboard does not implement it yet. For now, you can use install [Konele](https://github.com/Kaljurand/K6nele), and choose Sayboard as the backend.

Thanks for the input! Sayboard already has a similar feature: if you swipe left from the backspace key, text will be selected and then deleted when you lift your finger

Implemented in 14a815a9bc33d2dd7814b1b6d67d477159dddbfc

This is likely because of the new services permission in android 14. I'll need to increase the API target.

The Speech to text engine is vosk: https://alphacephei.com/vosk/ As far as I understand it does have a fixed word list, but simply updating it isn't enough - you have to...

The underlined words are what Vosk currently detects from you speech, but isn't sure about yet. You'll notice that the underlined words are changed while you speak. These words are...

You can try installing the large models manually: https://github.com/ElishaAz/Sayboard/wiki/Install-a-Vosk-Model-Manually

Which version of Florisboard? What android version?