Speech Denoising Python Packages

resemble-enhance

AI powered speech denoising and enhancement

torch-log-wmse

logWMSE, an audio quality metric & loss function with support for digital silence target. Useful for training and evaluating audio source separation systems.

torch-l1-snr

Variations of L1 SNR Loss function for training audio source separation machine learning models

dpdfnet

DPDFNet: causal single-channel speech enhancement that boosts DeepFilterNet2 with dual-path RNN blocks for stronger long-range temporal and cross-band modeling. Repo includes PyTorch implementation + checkpoints, ONNX & TFLite models with inference code, and a real-time demo.

moises-light

Unofficial PyTorch implementation of "Moises-Light: Resource-efficient Band-split U-Net For Music Source Separation"

torch-log-wmse-audio-quality

logWMSE, an audio quality metric & loss function with support for digital silence target. Useful for training and evaluating audio source separation systems.

rope-conformer

A clean, plug-and-play Conformer encoder with rotary positional embeddings.