The PyTorch-based audio source separation toolkit for researchers

Unofficial PyTorch implementation of Google AI's VoiceFilter system

A PyTorch implementation of Conv-TasNet described in "TasNet: Surpassing Ideal Time-Frequency Masking for Speech Separation" with Permutation Invariant Training (PIT).

Deep Recurrent Neural Networks for Source Separation

Two-talker Speech Separation with LSTM/BLSTM by Permutation Invariant Training method.

A PyTorch implementation of DNN-based source separation.

A PyTorch implementation of Time-domain Audio Separation Network (TasNet) with Permutation Invariant Training (PIT) for speech separation.

An implementation of audio source separation tools.

Adaptive and Focusing Neural Layers for Multi-Speaker Separation Problem

Ultimate Vocal Remover for Google Colab

Real time multilingual face translator

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

Software that performs the separation of vocals from music using neural networks (part of my Bachelor's thesis).

Code and datasets for 'Move2Hear: Active Audio-Visual Source Separation' (ICCV 2021)

A convolutional neural network for blind audio source separation.

An exploration of blind source audio separation using spiking neural networks. Latency, power. and intelligibility are primary objectives while bio-plausibility is left as a secondary objective to be addressed in the future.

eCMU: An Efficient Phase-aware Framework for Music Source Separation with Conformer (IEEE RIVF23)

OpenVINO DevCUP music aeparation & transcription

Download multiple tracks from youtube by a single query - with GUI.

Deezer source separation library including pretrained models.