Google voice foss replica!:)

Is a good project. Can it be used in keyboard in phones!

2 Likes

Yes, check this APP:

1 Like

If I’m not mistaken, that’s what Gadgetbridge is about to implement – and what would enable dictation via e.g. a Pebble smart watch.

I am currently playing with https://github.com/alphacep/vosk-android-demo which uses a non standard maven repository

https://alphacephei.com/maven/

Is this a nogo for f-droid?

has somebody succeeded to build it with jitpack.io ?

I tried with

implementation 'com.github.alphacephei:vosk-api:0.3.23'

but it did not work

Afraid that maven repo will be a show-stopper, yes. You could try the submodule approach – or srclib.

There is maven central.

Why exactly is it a showstopper? Since jcenter is offline we moved to our repo which is much easier for us.

We have no problem to put jars anywhere else if you have good candidates.

Maybe for you – but for our builds, only trusted maven repos are permitted. And “putting JARs” won’t help, as F-Droid builds everything from source: JARs (or other blobs) are not permitted, unless from a trusted Maven repo (where they are built reproducibly).

Ok, we can put the build on jitpack.io I believe coming days.

@k3b we also have an app, not just a demo: https://github.com/alphacep/vosk-android-service, that app implements ASR service compatible with Android and gonna be a better candidate for F-droid. Please check. We will add model download/UI there soon so it will be a drop-in replacement for voice services.

1 Like

One can subscribe to https://github.com/alphacep/vosk-api/issues/558 to get jitpack updates

1 Like