Oren Amsalem

maybe remove the `try expect` also do you see the `load time..` print?

I think @jastern33 is correct, this is the result from this github:  And this is from - http://nlp.seas.harvard.edu/2018/04/03/attention.html 

Hi, Change this line: https://github.com/facebookresearch/av_hubert/blob/258fb50e155134eec2c4b49c2ae8de267075fd18/avhubert/hubert_dataset.py#L191 To this: ```` self.noise_wav, self.noise_prob, self.noise_snr, self.noise_num = [], noise_prob, noise_snr, noise_num ```

``` git clone https://github.com/zacharydenton/wavebender.git cd wavebender python setup.py install ```

any update on this one? Thanks

This one: https://github.com/ChrisBeaumont/mpl-modest-image/pull/12 Solves your issue.

Thanks @sanchit-gandhi, but what about the original question? is there a way to make sure Whisper outputs numbers with letter and not number tokens? E.g. if the audio contains someone...

Hi all, if you look at the github commits https://github.com/facebookresearch/fairseq2/commits/main/ you can see that they are working on it, I am not sure constant question about it is useful (@cbalioglu...

yeah, it is an error in the instructions. See here: https://github.com/facebookresearch/jepa/pull/35#issue-2172512717 Try configs/evals/vitl16_k400.yaml. (with an s)

@pzelasko what pytorch version did you use? We found that the latest versions where more stable. Did you try with bf16_mixed?