meronym / speaker-diarization Goto Github PK

About

First of all, thank you so much for publishing your code. It's great to see some good examples where features are seperated into files, and not just one huge scoring/prediction script. ❤️

In predict.py you reference an s3 link, and I've been trying to figure out who it's owned by. Is it a pyannote.audio owned mirror, or are you hosting it? The reason for me asking, is to make sure that I keep an eye on it, when an update comes to the models. ✌🏻

Have a good one, and thanks again!

When running the most recent version of the diarization model I get this error on every .wav and .m4a file that I try. It seems to work fine on the second most recent version however.

Most recent: https://replicate.com/meronym/speaker-diarization/versions/64b78c82f74d78164b49178443c819445f5dca2c51c8ec374783d49382342119

Second most recent: https://replicate.com/meronym/speaker-diarization/versions/f9e76a15879810d34a6707e01def0150913db23fda77e18664c789501dc08ec5

Running predict()...
pre-processing audio file...
/tmp/tmpmkvim7bx55c6f818-aa83-44ff-bf29-6391baddabb0: Invalid data found when processing input