More speech-to-text models for dictation purposes

[Mk-N]

Currently only OpenAI whisper models can be chosen. Furthermore, the turbo whisper model cannot be chosen and it is unclear whether the large model refers to large v1, v2 or v3.

The ability to have a greater choice of models from different companies would be a nice to have.

[Jeffser]

The list of models come from this part of the readme

https://github.com/openai/whisper?tab=readme-ov-file#available-models-and-languages

[linuxkernel94]

Hi Jeffser, I don't see German on Alpaca, is it a bug?

Thank you!

[linuxkernel94]

Edit: I don't see it on text to speech, I can see it on speech to text 😅

[Jeffser]

Hi @linuxkernel94 unfortunately Kokoro, the modelset I'm using for voices doesn't have any voices in German.

I also don't include Japanese and Chinese because it requires additional libraries that are way too big