Great tool, only marred by Microsoft Speech Recognition. I don't understand why MS cannot produce better speech recognition for Windows.
Keep up the great work.
Did you consider adding support for switching to other speech recognition libraries by abstracting access to those features? You can have adaption source code between your code and the speech recognition library. In which case your source code can be kept closed.
This one may be interesting:
https://github.com/ggerganov/whisper.cpp?tab=readme-ov-file
MS voice recognition is a pain even for a native speaker. For a not native speaker like me this software(and alternatives like VA) is COMPLETELY useless.
So, alternative engine is a must.
Is there an option to use Google recognition API? And feed data via TCP from API? Everyone who wants to pay for google engine can do it on their own, just pasting API key in app
Currently no. I am looking into a potential free alternative engine first, primarily to support Linux, before considering any paid engine support. I don't make anything doing this, aside from the few kindly donations, and I don't particularly feel like promoting someone who does, but we'll see in the future. Aside from voice control, there are those using it for other functionality such as a replacement for JoyToKey and Joystick Gremlin as well, so although perhaps not for you, there is value in it for others.
