whisper supports auto-detection of spoken language, but also supports manual configuration at runtime (https://github.com/openai/whisper/blob/ba3f3cd54b0e5b8ce1ab3de13e32122d0d5f98ab/whisper/tokenizer.py#L10). From our anecdotal experience, the auto-detection accuracy is not quite satisfying, and AAPB users often know what the primary language used in a collection, we'd like to support the language picker as a runtime parameter in this wrapper as well.
New Feature Summary
whisper supports auto-detection of spoken language, but also supports manual configuration at runtime (https://github.com/openai/whisper/blob/ba3f3cd54b0e5b8ce1ab3de13e32122d0d5f98ab/whisper/tokenizer.py#L10). From our anecdotal experience, the auto-detection accuracy is not quite satisfying, and AAPB users often know what the primary language used in a collection, we'd like to support the language picker as a runtime parameter in this wrapper as well.
Related
No response
Alternatives
No response
Additional context
No response