gpo.zugaina.org

Search Portage & Overlays:

sci-ml/torch-audiomentations

Fast, differentiable, GPU-friendly audio augmentations for PyTorch

Screenshots

  • torch-audiomentations-0.12.0
    ~amd64
    python_single_target_python3_12 python_single_target_python3_13 python_single_target_python3_14

    View      Download      Browse     License: MIT   
    Overlay: stuff

ChangeLog

commit d6384b109c3b359a84d694c3b620345f862b486a
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Mon May 18 19:47:01 2026 +0200

sci-ml/torch-audiomentations: follow upstream github move to iver56 fork

asteroid-team/torch-audiomentations permanently redirects to
iver56/torch-audiomentations (project maintained by the original
author after the asteroid-team org went dormant). Update HOMEPAGE
+ metadata.xml remote-id + bugs-to.

commit cb19fe0f5b416045ce5f17bb57cf331891c310eb
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Sat May 16 10:09:46 2026 +0200

sci-ml/torch-audiomentations: new package, add 0.12.0

Differentiable, GPU-friendly audio augmentations for PyTorch
(SpecAugment, room reverb, pitch shift, ...). Required runtime
dep of sci-ml/pyannote-audio for training-time augmentation.

SINGLE_IMPL via pytorch. PyYAML is upstream extras_require only
(lazy import in utils/config.py); downstream pyannote-audio
pulls dev-python/pyyaml directly so we omit it here.