Towards an Efficient Deep Learning Model for Emotion and Theme Recognition in Music

Srividya Tirunellai Rajamani, Kumar Rajamani, Bjorn W. Schuller

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

1 Scopus citations

Abstract

Emotion and theme recognition in music plays a vital role in music information retrieval and recommendation systems. Deep learning based techniques have shown great promise in this regard. Realising optimal network configurations with least number of floating point operations per second (FLOPS) and model parameters is of paramount importance to obtain efficient deployable models, especially for resource constrained hardware. We propose a novel integration of stand-alone self-attention into a Visual Geometry Group (VGG)-like network for the task of multi-label emotion and theme recognition in music. Through extensive experimental evaluation, we discover the ideal and optimal integration of stand-alone self-attention which leads to substantial reduction in number of parameters and FLOPS, yet yielding better performance. We benchmark our results on the autotagging-moodtheme subset of the MTG-Jamendo dataset. Using mel-spectrogram as the input, we demonstrate that our proposed SA-VGG network requires 55 % fewer parameters and 60 % fewer FLOPS while improving the baseline ROC-AUC and PR-AUC.

Original languageEnglish
Title of host publicationIEEE 23rd International Workshop on Multimedia Signal Processing, MMSP 2021
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781665432870
DOIs
StatePublished - 2021
Externally publishedYes
Event23rd IEEE International Workshop on Multimedia Signal Processing, MMSP 2021 - Tampere, Finland
Duration: 6 Oct 20218 Oct 2021

Publication series

NameIEEE 23rd International Workshop on Multimedia Signal Processing, MMSP 2021

Conference

Conference23rd IEEE International Workshop on Multimedia Signal Processing, MMSP 2021
Country/TerritoryFinland
CityTampere
Period6/10/218/10/21

Keywords

  • VGG
  • automatic music tagging
  • multi-label classification
  • music emotion recognition
  • self-attention

Fingerprint

Dive into the research topics of 'Towards an Efficient Deep Learning Model for Emotion and Theme Recognition in Music'. Together they form a unique fingerprint.

Cite this