C++17 port of Open-Unmix-PyTorch with streaming LSTM inference, ggml, quantization, and Eigen
-
Updated
Mar 15, 2025 - C++
C++17 port of Open-Unmix-PyTorch with streaming LSTM inference, ggml, quantization, and Eigen
Pytorch implementation of MDensenet and sparse NMF. Made for my undergraduate thesis "Music Source Separation with Supervised Learning Methods".
A website to demonstrate the working of Open Unmix model where user uploads a .mp3 / .wav song file and gets 10sec vocals + instruments results generated from the uploaded file using the model.
Stem Player — local-first ML stem separation & remix console: split any track into vocals/drums/bass/other on your own GPU, then mix and export in the browser. Public showcase of functionality, ML methodology, and architecture (not the full source).
🎧 Peel — split any song into vocals, drums, bass & instruments. 100% in your browser, free, private. No upload, no account.
Clojure bindings for PyTorch implementation of Open-Unmix
Add a description, image, and links to the open-unmix topic page so that developers can more easily learn about it.
To associate your repository with the open-unmix topic, visit your repo's landing page and select "manage topics."