Depthwise Separable Convolutions Versus Recurrent Neural Networks for Monaural Singing Voice Separation

Pyry Pyykkönen, Stylianos I. Mimilakis, Konstantinos Drossos, Tuomas Virtanen

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

3 Citations (Scopus)

Abstract

Recent approaches for music source separation are almost exclusively based on deep neural networks, mostly employing recurrent neural networks (RNNs). Although RNNs are in many cases superior than other types of deep neural networks for sequence processing, they are known to have specific difficulties in training and parallelization, especially for the typically long sequences encountered in music source separation. In this paper we present a use-case of replacing RNNs with depth-wise separable (DWS) convolutions, which are a lightweight and faster variant of the typical convolutions. We focus on singing voice separation, employing an RNN architecture, and we replace the RNNs with DWS convolutions (DWS-CNNs). We conduct an ablation study and examine the effect of the number of channels and layers of DWS-CNNs on the source separation performance, by utilizing the standard metrics of signal-to-artifacts, signal-to-interference, and signal-to-distortion ratio. Our results show that by replacing RNNs with DWS-CNNs yields an improvement of 1.20, 0.06, 0.37 dB, respectively, while using only 20.57% of the amount of parameters of the RNN architecture.
Original languageEnglish
Title of host publicationIEEE International Workshop on Multimedia Signal Processing (MMSP)
Number of pages6
ISBN (Electronic)978-1-7281-9320-5
DOIs
Publication statusPublished - 2020
Publication typeA4 Article in conference proceedings
EventIEEE International Workshop on Multimedia Signal Processing -
Duration: 1 Jan 1900 → …

Publication series

NameIEEE International Workshop on Multimedia Signal Processing
ISSN (Electronic)2473-3628

Conference

ConferenceIEEE International Workshop on Multimedia Signal Processing
Period1/01/00 → …

Publication forum classification

  • Publication forum level 1

Fingerprint

Dive into the research topics of 'Depthwise Separable Convolutions Versus Recurrent Neural Networks for Monaural Singing Voice Separation'. Together they form a unique fingerprint.

Cite this