Hugging Face's logo

Spaces:
ConvxO2
/
Who-Spoke-When
Running

App Files Community
Who-Spoke-When / models

Commit History

  1. Reduce speaker over-segmentation in auto clustering Running

    ConvxO2
    789006e

  1. Add pyannote-first diarization path and tune fallback clustering

    ConvxO2
    8d04859

  2. Fix stereo duration handling and robust ECAPA model loading

    ConvxO2
    4b8c370

  3. Fix portability, model cache handling, and deploy token safety

    ConvxO2
    f54b658

  1. Fix LocalStrategy.COPY for speechbrain 1.0.3

    ConvxO2
    7ff62ef

  2. Fix from_hparams for speechbrain 1.0.3

    ConvxO2
    76debc2

  3. Fix cache dir for HuggingFace Spaces

    ConvxO2
    c16fd7c

  4. Initial commit: Speaker Diarization System

    ConvxO2
    d7a2919