crlandsc/torch-l1-snr
Variations of L1 SNR Loss function for training audio source separation machine learning models
This package helps audio engineers and researchers improve the quality of machine learning models for audio source separation. It takes a model's estimated audio components (like vocals or instruments) and the actual, isolated components, then provides a score that helps the model learn to produce clearer separations. This is useful for anyone training models to untangle mixed audio, such as for music production, speech enhancement, or environmental sound analysis.
Available on PyPI.
Use this if you are developing or training machine learning models that need to accurately separate mixed audio signals into their constituent parts.
Not ideal if you are an end-user simply looking to separate audio files without needing to train a custom machine learning model.
Stars
43
Forks
—
Language
Python
License
MIT
Category
Last pushed
Feb 24, 2026
Commits (30d)
0
Dependencies
3
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/crlandsc/torch-l1-snr"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
iver56/audiomentations
A Python library for audio data augmentation. Useful for making audio ML models work well in the...
Rikorose/DeepFilterNet
Noise supression using deep filtering
torchsynth/torchsynth
A GPU-optional modular synthesizer in pytorch, 16200x faster than realtime, for audio ML researchers.
marl/openl3
OpenL3: Open-source deep audio and image embeddings
archinetai/audio-data-pytorch
A collection of useful audio datasets and transforms for PyTorch.