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

TL;DR

This paper introduces a neural audio codec based on residual quantized variational autoencoders for musical source separation, achieving near state-of-the-art results with reduced computational requirements.

Contribution

The paper presents a novel residual quantized variational autoencoder architecture for source separation, trained on multi-track music data, with improved efficiency and competitive performance.

Findings

Achieves near state-of-the-art separation results

Requires less computational power than comparable models

Code is publicly available for reproducibility

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