More than 1 default language in settings for better speech recognition for bilingual users

[it-sha]

Typical workflow: If English is not your main language, you have to switch between two languages a lot of times.
Auto-detection for these cases can provide much worse results than a predefined selection of two languages. For example, German or English, Ukrainian or English, and so on.

It's possible to do this with Wispr Flow, and it works perfectly
It would be amazing if we were able to set it with a Handy as well.

[cjpais]

In the past I've said no to this request but I do believe it's possible with Whisper however not the other models as far as I'm aware

It's fairly low priority at the moment, but if someone wants to submit a PR to transcribe rs for this I will definitely pull it in

[AlexanderYastrebov]

Handy could also keep track of the current input layout or have a second hotkey and then use different models for diffrent languages.

[cjpais]

This is actually a PR already, if you want to do a review of the code @AlexanderYastrebov that would be nice. I've mostly been pulling in fixes rather than features recently so it's not been top priority