SMERS: Music emotion recognition using support vector regression

Byeong Jun Han, Seungmin Rho, Roger B. Dannenberg, Een Jun Hwang

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

60 Citations (Scopus)

Abstract

Music emotion plays an important role in music retrieval, mood detection and other music-related applications. Many issues for music emotion recognition have been addressed by different disciplines such as physiology, psychology, cognitive science and musicology. We present a support vector regression (SVR) based music emotion recognition system. The recognition process consists of three steps: (i) seven distinct features are extracted from music; (ii) those features are mapped into eleven emotion categories on Thayer's two-dimensional emotion model; (iii) two regression functions are trained using SVR and then arousal and valence values are predicted. We have tested our SVR-based emotion classifier in both Cartesian and polar coordinate system empirically. The result indicates the SVR classifier in the polar representation produces satisfactory result which reaches 94.55% accuracy superior to the SVR (in Cartesian) and other machine learning classification algorithms such as SVM and GMM.

Original languageEnglish
Title of host publicationProceedings of the 10th International Society for Music Information Retrieval Conference, ISMIR 2009
Pages651-656
Number of pages6
Publication statusPublished - 2009 Dec 1
Event10th International Society for Music Information Retrieval Conference, ISMIR 2009 - Kobe, Japan
Duration: 2009 Oct 262009 Oct 30

Other

Other10th International Society for Music Information Retrieval Conference, ISMIR 2009
CountryJapan
CityKobe
Period09/10/2609/10/30

Fingerprint

Classifiers
Physiology
Learning systems
Music
Emotion Recognition
Emotion
Classifier
Cartesian
Mood
Cognitive Science
Psychology
Musicology
Machine Learning
Valence
Arousal

ASJC Scopus subject areas

  • Music
  • Information Systems

Cite this

Han, B. J., Rho, S., Dannenberg, R. B., & Hwang, E. J. (2009). SMERS: Music emotion recognition using support vector regression. In Proceedings of the 10th International Society for Music Information Retrieval Conference, ISMIR 2009 (pp. 651-656)

SMERS : Music emotion recognition using support vector regression. / Han, Byeong Jun; Rho, Seungmin; Dannenberg, Roger B.; Hwang, Een Jun.

Proceedings of the 10th International Society for Music Information Retrieval Conference, ISMIR 2009. 2009. p. 651-656.

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

Han, BJ, Rho, S, Dannenberg, RB & Hwang, EJ 2009, SMERS: Music emotion recognition using support vector regression. in Proceedings of the 10th International Society for Music Information Retrieval Conference, ISMIR 2009. pp. 651-656, 10th International Society for Music Information Retrieval Conference, ISMIR 2009, Kobe, Japan, 09/10/26.
Han BJ, Rho S, Dannenberg RB, Hwang EJ. SMERS: Music emotion recognition using support vector regression. In Proceedings of the 10th International Society for Music Information Retrieval Conference, ISMIR 2009. 2009. p. 651-656
Han, Byeong Jun ; Rho, Seungmin ; Dannenberg, Roger B. ; Hwang, Een Jun. / SMERS : Music emotion recognition using support vector regression. Proceedings of the 10th International Society for Music Information Retrieval Conference, ISMIR 2009. 2009. pp. 651-656
@inproceedings{077f13a3cadf48ecb8daa6e3c8c5fe16,
title = "SMERS: Music emotion recognition using support vector regression",
abstract = "Music emotion plays an important role in music retrieval, mood detection and other music-related applications. Many issues for music emotion recognition have been addressed by different disciplines such as physiology, psychology, cognitive science and musicology. We present a support vector regression (SVR) based music emotion recognition system. The recognition process consists of three steps: (i) seven distinct features are extracted from music; (ii) those features are mapped into eleven emotion categories on Thayer's two-dimensional emotion model; (iii) two regression functions are trained using SVR and then arousal and valence values are predicted. We have tested our SVR-based emotion classifier in both Cartesian and polar coordinate system empirically. The result indicates the SVR classifier in the polar representation produces satisfactory result which reaches 94.55{\%} accuracy superior to the SVR (in Cartesian) and other machine learning classification algorithms such as SVM and GMM.",
author = "Han, {Byeong Jun} and Seungmin Rho and Dannenberg, {Roger B.} and Hwang, {Een Jun}",
year = "2009",
month = "12",
day = "1",
language = "English",
isbn = "9780981353708",
pages = "651--656",
booktitle = "Proceedings of the 10th International Society for Music Information Retrieval Conference, ISMIR 2009",

}

TY - GEN

T1 - SMERS

T2 - Music emotion recognition using support vector regression

AU - Han, Byeong Jun

AU - Rho, Seungmin

AU - Dannenberg, Roger B.

AU - Hwang, Een Jun

PY - 2009/12/1

Y1 - 2009/12/1

N2 - Music emotion plays an important role in music retrieval, mood detection and other music-related applications. Many issues for music emotion recognition have been addressed by different disciplines such as physiology, psychology, cognitive science and musicology. We present a support vector regression (SVR) based music emotion recognition system. The recognition process consists of three steps: (i) seven distinct features are extracted from music; (ii) those features are mapped into eleven emotion categories on Thayer's two-dimensional emotion model; (iii) two regression functions are trained using SVR and then arousal and valence values are predicted. We have tested our SVR-based emotion classifier in both Cartesian and polar coordinate system empirically. The result indicates the SVR classifier in the polar representation produces satisfactory result which reaches 94.55% accuracy superior to the SVR (in Cartesian) and other machine learning classification algorithms such as SVM and GMM.

AB - Music emotion plays an important role in music retrieval, mood detection and other music-related applications. Many issues for music emotion recognition have been addressed by different disciplines such as physiology, psychology, cognitive science and musicology. We present a support vector regression (SVR) based music emotion recognition system. The recognition process consists of three steps: (i) seven distinct features are extracted from music; (ii) those features are mapped into eleven emotion categories on Thayer's two-dimensional emotion model; (iii) two regression functions are trained using SVR and then arousal and valence values are predicted. We have tested our SVR-based emotion classifier in both Cartesian and polar coordinate system empirically. The result indicates the SVR classifier in the polar representation produces satisfactory result which reaches 94.55% accuracy superior to the SVR (in Cartesian) and other machine learning classification algorithms such as SVM and GMM.

UR - http://www.scopus.com/inward/record.url?scp=84873695101&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84873695101&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84873695101

SN - 9780981353708

SP - 651

EP - 656

BT - Proceedings of the 10th International Society for Music Information Retrieval Conference, ISMIR 2009

ER -