Did any hugging face permissions change with the release of 3.0?

[maxchehab]

I'm still using 2.1.1 and this morning around 1am PST started to get Huggingface permission errors.

Pipeline.from_pretrained('pyannote/speaker-diarization', use_auth_token=HF_TOKEN)

Could not download 'pyannote/speaker-diarization' pipeline.
It might be because the pipeline is private or gated so make
sure to authenticate. Visit https://hf.co/settings/tokens to
create your access token and retry with:

   >>> Pipeline.from_pretrained('pyannote/speaker-diarization',
   ...                          use_auth_token=YOUR_AUTH_TOKEN)

If this still does not work, it might be because the pipeline is gated:
visit https://hf.co/pyannote/speaker-diarization to accept the user conditions.

I've accepted the user conditions for both the segmentation and speaker-diarization models and the model was working fine before 1am PST.

Is it possible that permissions were changed in Huggingface that would impact 2.1.1 users?

[github-actions]

Thank you for your issue.
We found the following entry in the FAQ which you may find helpful:

• Can I use gated models (and pipelines) offline?

Feel free to close this issue if you found an answer in the FAQ.

If your issue is a feature request, please read this first and update your request accordingly, if needed.

If your issue is a bug report, please provide a minimum reproducible example as a link to a self-contained Google Colab notebook containing everthing needed to reproduce the bug:

• installation
• data preparation
• model download
• etc.

Providing an MRE will increase your chance of getting an answer from the community (either maintainers or other power users).

Companies relying on pyannote.audio in production may contact me via email regarding:

• paid scientific consulting around speaker diarization and speech processing in general;
• custom models and tailored features (via the local tech transfer office).

This is an automated reply, generated by FAQtory

[hbredin]

That is strange. Can you please try again?

[maxchehab]

I have been all morning :)

I've also tried to create a new hugging face account, accept the user conditions and same result.

[hbredin]

At some point, I renamed pyannote/speaker-diarization to pyannote/speaker-diarization-2.1, thinking that Huggingface would also forward the authorization.

So I have just renamed it back to pyannote/speaker-diarization.

cc @julien-c are you aware of this problem with gated models? when you rename it, authorizations seem to be removed?

[maxchehab]

Confirmed the pipeline is working now - first successful download at 9:05 am PST.

I suggest we keep the name for now to support 2.1.1

[remic33]

Seems to be working now, thanks @hbredin

[Dannypeja]

Weirdly I am having this issue when I use whisperX inside Docker.
A few days ago it was working fine, and I think I did not break it..

Could not download 'pyannote/speaker-diarization-3.0' pipeline.
It might be because the pipeline is private or gated so make
sure to authenticate. Visit https://hf.co/settings/tokens to
create your access token and retry with:

   >>> Pipeline.from_pretrained('pyannote/speaker-diarization-3.0',
   ...                          use_auth_token=YOUR_AUTH_TOKEN)

[remic33]

Hi, there were somes changes with the pyannote update.
You may have to accept terms for version 3.
Can find them there :
https://huggingface.co/pyannote/speaker-diarization-3.0