Hervé BREDIN comments

Results 270 comments of


                                            Hervé BREDIN

numpy.ndarray audio input doesn't work?

Thanks for the bug report. Would you mind opening a PR removing the mention of numpy arrays in the error message?

numpy.ndarray audio input doesn't work?

Thanks for the PR. Please open a new issue/discussion for your other question.

Update Pyannote with SpeechBrain 1.0

Thanks for the proactive fix. I'd rather switch to `speechbrain 1.x` directly. Could you update the PR accordingly (both `requirements.txt` and `speaker_verification.py`?

Update Pyannote with SpeechBrain 1.0

To detect future breaking changes, would you mind adding a unit test that simply loads speechbrain ECAPA model? and then I think it should be good.

Update Pyannote with SpeechBrain 1.0

Looks like some tests are failing.

Update Pyannote with SpeechBrain 1.0

What about waiting for the fix to end up in a new release? That being said, Python 3.8 apparently only represents 3% of pyannote userbase. I might get rid of...

Update Pyannote with SpeechBrain 1.0

Great. Let me know when it is released and I will update the `requirements.txt`.

Update Pyannote with SpeechBrain 1.0

Thanks @Adel-Moumen! Will be part of next release!

pyannote uninstalling torch cuda and install torch cpu

I'd suggest using the same `--index-url` when installing `pyannote.audio`. ```bash pip install pyannote.audio --index-url https://download.pytorch.org/whl/cu118 ```

pyannote uninstalling torch cuda and install torch cpu

It would be cool if you could open a PR fixing this now that you have found a solution.