Hervé BREDIN comments

Results 270 comments of


                                            Hervé BREDIN

Cannot instantiate parameters on finetuned model

Please update the MRE to also take care of installing dependencies. It fails at the first cell.

fix: fix #1709

Merged in `develop`. Will be part of next release.

Wrong usage of meta-protocols subsets in segmentation tasks

@clement-pages any idea?

Evaluating speaker diarization systems with γ inter-annotator agreement

Thank you both for your detailed answers. To summarize my understanding: using this metric for speaker diarization is not that obvious and remains an open research question. Thinking out loud:...

memory optimizations for pyannote.audio.core.inference.Inference.aggregate()

Merged! 🎉 Thanks a lot for your contribution. Will be part of next release.

memory optimizations for pyannote.audio.core.inference.Inference.aggregate()

I'd love to know more about how pyannote impacts your work. Feel free to drop me an email!

Speakers with similar pitch are difficult to distinguish

As for any machine learning approaches, train/test domain mismatch is usually the culprit. Fine-tuning internal models and pipelines to your use case data is usually the best solution. Did you...

tuto: add `eval_seriously_pipeline` tutorial

FYI, @joonaskalda's PR has been merged.

Error while optimizing hyperparameters of clustering pipeline

Thanks for the report. Would you mind opening a PR adding this `import` back?

Update to latest optuna

`FutureWarning: suggest_uniform has been deprecated in v3.0.0. This feature will be removed in v6.0.0. See https://github.com/optuna/optuna/releases/tag/v3.0.0. Use suggest_float instead`