Benchmarks and leaderboards for sound demixing tasks

TL;DR

This paper introduces new benchmarks and leaderboards for sound demixing, compares popular models, and proposes an ensemble approach that achieved top results in the Music Demixing Challenge 2023.

Contribution

It provides new benchmark datasets, a comprehensive comparison of models, and a novel ensemble method for improved sound source separation.

Findings

New benchmark datasets for sound demixing

Ensemble approach outperforms individual models

Achieved top results in Music Demixing Challenge 2023

Abstract

Music demixing is the task of separating different tracks from the given single audio signal into components, such as drums, bass, and vocals from the rest of the accompaniment. Separation of sources is useful for a range of areas, including entertainment and hearing aids. In this paper, we introduce two new benchmarks for the sound source separation tasks and compare popular models for sound demixing, as well as their ensembles, on these benchmarks. For the models' assessments, we provide the leaderboard at https://mvsep.com/quality_checker/, giving a comparison for a range of models. The new benchmark datasets are available for download. We also develop a novel approach for audio separation, based on the ensembling of different models that are suited best for the particular stem. The proposed solution was evaluated in the context of the Music Demixing Challenge 2023 and achieved top results in different tracks of the challenge. The code and the approach are open-sourced on GitHub.