Support for VibeVoice

[FerLuisxd]

I would like to see support for
https://huggingface.co/microsoft/VibeVoice-ASR

As the model supports speaker diarization and auto language detection

[cjpais]

I actually have a bit of an implementation already working for it, but I will complete it and bring it in

Mostly it's been waiting for the other diarization changes to make it in.