tasnet

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

A PyTorch implementation of DNN-based source separation.

A PyTorch implementation of "TasNet: Surpassing Ideal Time-Frequency Masking for Speech Separation" (see recipes in aps framework https://github.com/funcwj/aps)

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

The state-of-art time domain network for speech separation, and it performs well on speech enhancement and music separation

Time-domain Audio Separation Network (IN PYTORCH)

Tensorflow 2.x (with Keras API) Implementation of the TaSNet (Luo et al., 2018)

A software-defined AI system for maritime noise reduction using a 3-stage deep learning pipeline (YAMNet, CRNN, TasNet). Achieved 98% classification accuracy and 60% noise reduction on the QiandaoEar22 dataset. Deployed as an offline-capable web app using Flask and SQLite.

Self-hostable web app for isolating the vocal, accompaniment, bass, and drums of any song. Supports Spleeter, D3Net, Demucs, Tasnet, X-UMX. Built with React and Django.

A PyTorch implementation of "Improving noise robust automatic speech recognition with single-channel time-domain enhancement network"