Source Separation of Multi-source Raw Music using a Residual Quantized Variational Autoencoder

Leonardo Berti
8/12/2024
cs.SDcs.LGcs.MM

No official code found. Best implementation: LeonardoBerti00/Source-Separation-of-Multi-source-Music-using-Residual-Quantizad-Variational-Autoencoder (6 stars, MIT, last commit 2026-01-17).

Abstract

I developed a neural audio codec model based on the residual quantized variational autoencoder architecture. I train the model on the Slakh2100 dataset, a standard dataset for musical source separation, composed of multi-track audio. The model can separate audio sources, achieving almost SoTA results with much less computing power. The code is publicly available at github.com/LeonardoBerti00/Source-Separation-of-Multi-source-Music-using-Residual-Quantizad-Variational-Autoencoder

View on arXivView PDF

Code Implementations(1)

LeonardoBerti00/Source-Separation-of-Multi-source-Music-using-Residual-Quantizad-Variational-AutoencoderCommunity100%
MIT

Source Separation of Multi-source Raw Music using a Residual Quantized variational Autoencoder

61PythonSep 14, 20239 months agoMIT
audioautoregressivedeep-learningresidual-quantizationsource-separation

Similar Papers

Learning Source Disentanglement in Neural Audio Codec

83% match1 repo482024

SNAC: Multi-Scale Neural Audio Codec

81% match3 repos7742024

A Conditioned UNet for Music Source Separation

80% match2025

Mamba2 Meets Silence: Robust Vocal Source Separation for Sparse Regions

79% match2025

Source Separation & Automatic Transcription for Music

79% match1 repo182024

Unsupervised Composable Representations for Audio

79% match1 repo142024

Cite this paper

@article{berti2024source,
  title  = {Source Separation of Multi-source Raw Music using a Residual Quantized Variational Autoencoder},
  author = {Leonardo Berti},
  year   = {2024},
  eprint = {2408.07020},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2408.07020v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2408.07020.svg)](https://codeofpaper.com/paper/2408.07020)

Discussion