VibeVoice-ASR

[Lunat1q]

There is this relatively new voice recognition model: https://github.com/microsoft/VibeVoice/tree/main
I've tried to use it with Buzz but no luck, is there a chance someone managed to make it work?

[raivisdejus]

My main question is if it is any good. If it is faster or more accurate than the models already supported.
If so we can look at implementing it, but some real world testing would be needed.

[Lunat1q]

I did several tests on it yesterday and today, I think it's significantly more accurate than Whisper LargeV3, and speaker identification is very precise compared to what buzz supports now.
I did test in several languages and in non-English I see the biggest improvement.