Add foreign language speech to text models in gallery

[SuperPat45]

Is your feature request related to a problem? Please describe.
The whisper model available in gallery does nor work well for non-English language.

Describe the solution you'd like
It would be useful to have foreign language Speech To Text models in the gallery.

For French language, I found in HungingFace:
https://huggingface.co/bofenghuang/whisper-large-v3-french
https://huggingface.co/personalizedrefrigerator/whisper-tiny-fr

Describe alternatives you've considered

Additional context

[BjoKaSH]

I'd like to second this 😃 . Although for the German language.

Kidding aside, I think it would be great to have options for all the common European languages in the Model Gallery, at least for those for which free/open source models exist.

Unfortunately I don't know much about the requirements or process to add a model (i.e. write a proper config and what ever else is needed), otherwise I would try to setup some more STT models myself. Is there some good and beginners-level documentation how to do that? Although I am active in programming 30+ years, I am an absolute beginner on all these AI-Models stuff. The LocalAI documentation on how to add/configure models I have seen so far is completely incomprehensible to me (but last checked closely in fall 2024). I have simple no clue how all the parts in LocalAI work together and how the front-end, the back-ends and the GBs of tensor coefficients connect to do their "magic".