Aarhus University Seal / Aarhus Universitets segl

Peter Ahrendt

Co-occurrence models in music genre classification

Research output: Contribution to book/anthology/report/proceedingBook chapterResearchpeer-review

  • P. Ahrendt
  • J. Larsen, Technical University of Denmark
  • ,
  • Cyril Goutte, Department of Informatics and Mathematical Modeling, Denmark
Music genre classification has been investigated using many different methods, but most of them build on probabilistic models of feature vectors xr which only represent the short time segment with index r of the song. Here, three different co-occurrence models are proposed which instead consider the whole song as an integrated part of the probabilistic model. This was achieved by considering a song as a set of independent co-occurrences (s, xr ) (s is the song index) instead of just a set of independent (x r)'s. The models were tested against two baseline classification methods on a difficult 11 genre data set with a variety of modern music. The basis was a so-called AR feature representation of the music. Besides the benefit of having proper probabilistic models of the whole song, the lowest classification test errors were found using one of the proposed models
Original languageEnglish
Title of host publication2005 IEEE Workshop on Machine Learning for Signal Processing
Number of pages6
Publication year2005
ISBN (print)9780780395176
Publication statusPublished - 2005
Externally publishedYes
EventIEEE International Workshop on Machine Learning for Signal Processing - Mystic, Connecticut, United States
Duration: 28 Sep 200530 Sep 2005


ConferenceIEEE International Workshop on Machine Learning for Signal Processing
LandUnited States
ByMystic, Connecticut

See relations at Aarhus University Citationformats

ID: 70750477