Warning trained model and version problem

[PierreVannier]

Hey there

I've finally succeeded in installing on my M1 pro (that was a hustle). Now when I launch this code :

from pyannote.audio import Pipeline
pipeline = Pipeline.from_pretrained("pyannote/speaker-diarization@2.1",
                                    use_auth_token="ACCESS_TOKEN_GOES_HERE")

# apply the pipeline to an audio file
diarization = pipeline("audio.wav")

# dump the diarization output to disk using RTTM format
with open("audio.rttm", "w") as rttm:
    diarization.write_rttm(rttm)

I've got several warnings among them this one :

Model was trained with pyannote.audio 0.0.1, yours is 2.1.1. Bad things might happen unless you revert pyannote.audio to 0.x.
Model was trained with torch 1.10.0+cu102, yours is 2.0.0. Bad things might happen unless you revert torch to 1.x.

Is there anything I can do to "avoid" bad things ? I suspect I'd have to train model with the good version of pyannote and torch ? Thanks for your help.

[github-actions[bot]]

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. Otherwise, please give us a little time to review.

This is an automated reply, generated by FAQtory

[hbredin]

I guess you installed from develop branch. You can safely ignore those warnings for now.

[PierreVannier]

I guess you installed from develop branch. You can safely ignore those warnings for now.

Thanks @hbredin. I assume my M1 do not use GPU and that's why it takes a hell of a time to process a 45 minutes podcast with 3 speakers. I've been forced to kill the process after 2 hours doing nothing with the given script.

My wav file is a mono file in 16khz with the 3 speakers talking.

[yorozcogonzalez]

Hello. I am having exactly the same issue here in Ubuntu. I can ignore the warnings but it is running on the cpu. Have you found any solution? Thanks

[geekinchief22]

Is there a way to disable the warnings? Should I use a different branch?

[LiRem101]

I have the same issue and what irritates me about it is that as far as I got it "pyannote/speaker-diarization@2.1" seems to be done for 2.1.1 while "pyannote/speaker-diarization" has been done for 0.0.1. Can someone explain why the error still shows up?

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[nina-wwc]

Same problem here; installed with pip install pyannote.audio so it shouldn't be a dev branch issue. Plus other warnings (like complaining about torchvision when the figures are not even used). Is there a way to disable warnings? Too many unnecessary repetitions in the logs.

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.