Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

whisperkit-generate-model - not working #29

Open
omerarshad opened this issue Feb 14, 2025 · 4 comments
Open

whisperkit-generate-model - not working #29

omerarshad opened this issue Feb 14, 2025 · 4 comments

Comments

@omerarshad
Copy link

whisperkit-generate-model --model-version openai/whisper-tiny --disable-default-tests --output-dir ./

I used above command to convert a whisper model, but it is generating empty directory. Any leads?

@atiorh
Copy link
Contributor

atiorh commented Feb 14, 2025

@omerarshad Could you please remove --disable-default-tests and report back?

@omerarshad
Copy link
Author

getting error on: from tests import test_audio_encoder, test_text_decoder in generate_model.py. Was able to generate using generate_model.py independently then using: whisperkit-generate-model --model-version --output-dir

@omerarshad
Copy link
Author

@atiorh What if it not creating TextDecoder.mlmodel for a tarteel-ai/whisper-base-ar-quran model? where it does create for openai/whisper-tiny

@atiorh
Copy link
Contributor

atiorh commented Feb 17, 2025

I can help guide if you paste the error logs here. Given that openai/whisper-tiny works, It is likely that the checkpoint you are referring to does not conform to the original Whisper model repo structure

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants