There is a SayBoard app that makes Speech to text recognition with the same Vosk library. I have a small device kids watch with 480x480px display and only 1Gb or RAM. And now I have to install both apps and download the Vosk models twice. Maybe it would be possible to centralize this somehow or merge this two applications. This is probably out of scope but highly relevant and this would simplify usage and installation for users.
UPD found a related ElishaAz/Sayboard#6
There is a SayBoard app that makes Speech to text recognition with the same Vosk library. I have a small device kids watch with 480x480px display and only 1Gb or RAM. And now I have to install both apps and download the Vosk models twice. Maybe it would be possible to centralize this somehow or merge this two applications. This is probably out of scope but highly relevant and this would simplify usage and installation for users.
UPD found a related ElishaAz/Sayboard#6