Home Research Papers Datasets State of the Art Pricing

Discover, visualize, and connect AI research papers. Explore the latest trends and insights in artificial intelligence research.

Product

Home
Research Papers
About

Support

Contact
Terms of Service
Privacy Policy

© 2026 Papersgraph. All rights reserved.

Music Source Separation | State-of-the-Art

music-source-separation

Music Source Separation

3260 papers • 126 benchmarks • 313 datasets

Music source separation is the task of decomposing music into its constitutive components, e. g., yielding separated stems for the vocals, bass, and drums. ( Image credit: SigSep )

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in music-source-separation

Trend

Dataset

Best Model

Actions

MUSDB18

MUSDB18

MUSDB18-HQ

MUSDB18-HQ

Slakh2100

Slakh2100

Libraries

i

Use these libraries to find music-source-separation models and implementations

FaceOnLive/Spleeter-Android-iOS

4 papers 193

Datasets

MUSDB18

MedleyDB

Slakh2100

MIR-1K

MUSDB18-HQ

Subtasks

No subtasks available.

Most implemented papers

Wave-U-Net: A Multi-Scale Neural Network for End-to-End Audio Source Separation

D. Stoller, Sebastian Ewert, S. Dixon•Thu May 24 2018

The Wave-U-Net is proposed, an adaptation of the U-Net to the one-dimensional time domain, which repeatedly resamples feature maps to compute and combine features at different time scales and indicates that its architecture yields a performance comparable to a state-of-the-art spectrogram-based U- net architecture, given the same data.

665

Content

Introduction Benchmarks Datasets Subtasks Libraries Papers

CocoChorales

CocoChorales

MuseScore

0

Multi-Scale multi-band densenets for audio source separation

Yuki Mitsufuji, Naoya Takahashi•Wed Jun 28 2017

A novel network architecture that extends the recently developed densely connected convolutional network (DenseNet) and takes advantage of long contextual information and outperforms state-of-the-art results on SiSEC 2016 competition by a large margin in terms of signal-to-distortion ratio.

161 0

All For One And One For All: Improving Music Separation By Bridging Networks

Yuki Mitsufuji, Shusuke Takahashi, Ryosuke Sawata, S. Uhlich•Wed Oct 07 2020

Experimental results show that the performance of Open-Unmix (UMX), a well-known and state-of-the-art open-source library for music separation, can be improved by utilizing a multi-domain loss (MDL) and two combination schemes.

57 0

Music Source Separation With Band-Split RNN

Yi Luo, Jianwei Yu•Thu Sep 29 2022

Band-split RNN (BSRNN) is proposed, a frequency-domain model that explictly splits the spectrogram of the mixture into subbands and perform interleaved band-level and sequence-level modeling and a semi-supervised model finetuning pipeline that can further improve the performance of the model.

182 0

Adversarial Semi-Supervised Audio Source Separation Applied to Singing Voice Extraction

D. Stoller, Sebastian Ewert, S. Dixon•Mon Oct 30 2017

This work adopts adversarial training for music source separation with the aim of driving the separator towards outputs deemed as realistic by discriminator networks that are trained to tell apart real from separator samples.

73 0

SPLEETER: A FAST AND STATE-OF-THE ART MUSIC SOURCE SEPARATION TOOL WITH PRE-TRAINED MODELS

Romain Hennequin, Anis Khlif, Félix Voituret, Manuel Moussallam•Mon Dec 31 2018

Spleeter is a new tool for music source separation with pre-trained models based on Tensorflow that makes it possible to separate audio files into 2, 4 or 5 stems with a single command line using pre- trained models.

47 0

End-to-end music source separation: is it possible in the waveform domain?

Xavier Serra, Jordi Pons, Francesc Lluís•Sun Oct 28 2018

A Wavenet-based model is proposed and Wave-U-Net can outperform DeepConvSep, a recent spectrogram-based deep learning model, and the results confirm that waveform-based models can perform similarly (if not better) than a spectrogram/deep learning model.

75 0

Primal-dual algorithms for non-negative matrix factorization with the Kullback-Leibler divergence

F. Bach, F. Yanez•Wed Dec 03 2014

This paper proposes a first-order primal-dual algorithm for non-negative decomposition problems (one of the two factors is fixed) with the KL distance and provides an efficient heuristic way to select step-sizes.

21 0

Multi-channel U-Net for Music Source Separation

Emilia G'omez, Juan F. Montesinos, V. S. Kadandale, G. Haro•Sun Mar 22 2020

A fairly straightforward approach for music source separation is to train independent models, wherein each model is dedicated for estimating only a specific source. Training a single model to estimate multiple sources generally does not perform as well as the independent dedicated models. However, Conditioned U-Net (C-U-Net) uses a control mechanism to train a single model for multi-source separation and attempts to achieve a performance comparable to that of the dedicated models. We propose a multi-channel U-Net (M-U-Net) trained using a weighted multi-task loss as an alternative to the C-U-Net. We investigate two weighting strategies for our multi-task loss: 1) Dynamic Weighted Average (DWA), and 2) Energy Based Weighting (EBW). DWA determines the weights by tracking the rate of change of loss of each task during training. EBW aims to neutralize the effect of the training bias arising from the difference in energy levels of each of the sources in a mixture. Our methods provide three-fold advantages compared to C-U-Net: 1) Fewer effective training iterations per epoch, 2) Fewer trainable network parameters (no control parameters), and 3) Faster processing at inference. Our methods achieve performance comparable to that of C-U-Net and the dedicated U-Nets at a much lower training cost.

23 0

Adding a benchmark result helps the community track progress.