A comparison of audio signal preprocessing methods for deep neural networks on music tagging

Keunwoo Choi, György Fazekas, Mark Sandler, Kyunghyun Cho

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we empirically investigate the effect of audio preprocessing on music tagging with deep neural networks. We perform comprehensive experiments involving audio preprocessing using different time-frequency representations, logarithmic magnitude compression, frequency weighting, and scaling. We show that many commonly used input preprocessing techniques are redundant except magnitude compression.

Original languageEnglish (US)
Title of host publication2018 26th European Signal Processing Conference, EUSIPCO 2018
PublisherEuropean Signal Processing Conference, EUSIPCO
Pages1870-1874
Number of pages5
ISBN (Electronic)9789082797015
DOIs
StatePublished - Nov 29 2018
Event26th European Signal Processing Conference, EUSIPCO 2018 - Rome, Italy
Duration: Sep 3 2018Sep 7 2018

Publication series

NameEuropean Signal Processing Conference
Volume2018-September
ISSN (Print)2219-5491

Other

Other26th European Signal Processing Conference, EUSIPCO 2018
CountryItaly
CityRome
Period9/3/189/7/18

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'A comparison of audio signal preprocessing methods for deep neural networks on music tagging'. Together they form a unique fingerprint.

  • Cite this

    Choi, K., Fazekas, G., Sandler, M., & Cho, K. (2018). A comparison of audio signal preprocessing methods for deep neural networks on music tagging. In 2018 26th European Signal Processing Conference, EUSIPCO 2018 (pp. 1870-1874). [8553106] (European Signal Processing Conference; Vol. 2018-September). European Signal Processing Conference, EUSIPCO. https://doi.org/10.23919/EUSIPCO.2018.8553106