[FEAT] Support Multiple New Whisper Models?

[scyto]

Is this a new feature request?

• I have searched the existing issues

Wanted change

Whisper now supports new models not supported by the GPU tag image file.

Currently supported by this image file:
tiny, tiny.en, base, base.en, small, small.en, distil-small.en, medium, medium.en, distil-medium.en, large, large-v1, large-v2

Not supported by this image file:
large-v3, distil-large-v2, distil-large-v3, large-v3-turbo, or turbo

Please update to the version of faster-whisper that supported these models.

Reason for change

I would like to use larger models - especially v3-turbo with my nvidia GPUS, this should be more accurate and faster,

Proposed code change

Support the latest features?

https://github.com/SYSTRAN/faster-whisper/blob/a6f8fbae0060cccb8bbe71422d3546d8206ebfe1/faster_whisper/transcribe.py#L533

[github-actions]

Thanks for opening your first issue here! Be sure to follow the relevant issue templates, or risk having this issue marked as invalid.

[thespad]

The list in the readme is non-exhaustive (updating it every time they add support for new models is a slog) but if you provide an invalid model it will tell you and list all the currently valid ones:

expected one of: tiny.en, tiny, base.en, base, small.en, small, medium.en, medium, large-v1, large-v2, large-v3, large, distil-large-v2, distil-medium.en, distil-small.en

The turbo models haven't been included in an upstream release yet: https://github.com/SYSTRAN/faster-whisper/blob/v1.0.3/faster_whisper/transcribe.py#L102-L107