Back to Search Start Over

The whole is greater than the sum of its parts: improving music source separation by bridging networks

Authors :
Ryosuke Sawata
Naoya Takahashi
Stefan Uhlich
Shusuke Takahashi
Yuki Mitsufuji
Source :
EURASIP Journal on Audio, Speech, and Music Processing, Vol 2024, Iss 1, Pp 1-19 (2024)
Publication Year :
2024
Publisher :
SpringerOpen, 2024.

Abstract

Abstract This paper presents the crossing scheme (X-scheme) for improving the performance of deep neural network (DNN)-based music source separation (MSS) with almost no increasing calculation cost. It consists of three components: (i) multi-domain loss (MDL), (ii) bridging operation, which couples the individual instrument networks, and (iii) combination loss (CL). MDL enables the taking advantage of the frequency- and time-domain representations of audio signals. We modify the target network, i.e., the network architecture of the original DNN-based MSS, by adding bridging paths for each output instrument to share their information. MDL is then applied to the combinations of the output sources as well as each independent source; hence, we called it CL. MDL and CL can easily be applied to many DNN-based separation methods as they are merely loss functions that are only used during training and do not affect the inference step. Bridging operation does not increase the number of learnable parameters in the network. Experimental results showed that the validity of Open-Unmix (UMX), densely connected dilated DenseNet (D3Net) and convolutional time-domain audio separation network (Conv-TasNet) extended with our X-scheme, respectively called X-UMX, X-D3Net and X-Conv-TasNet, by comparing them with their original versions. We also verified the effectiveness of X-scheme in a large-scale data regime, showing its generality with respect to data size. X-UMX Large (X-UMXL), which was trained on large-scale internal data and used in our experiments, is newly available at https://github.com/asteroid-team/asteroid/tree/master/egs/musdb18/X-UMX .

Details

Language :
English
ISSN :
16874722
Volume :
2024
Issue :
1
Database :
Directory of Open Access Journals
Journal :
EURASIP Journal on Audio, Speech, and Music Processing
Publication Type :
Academic Journal
Accession number :
edsdoj.4cb69dcbfae44923b88de1070dc8e3a1
Document Type :
article
Full Text :
https://doi.org/10.1186/s13636-024-00354-6