Co-occurrence Models in Music Genre Classification

Peter Ahrendt, Cyril Goutte, Jan Larsen

AbstractMusic genre classification has been investigated using many different methods,
but most of them build on probabilistic models of feature vectors x_r 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, x_r) (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.
Keywordsmusic genre classification, probabilistic models
TypeConference paper [With referee]
ConferenceIEEE International workshop on Machine Learning for Signal Processing
EditorsV. Calhoun, T. Adali, J. Larsen, D. Miller, S. Douglas
Year2005    Month September    pp. 247-252
AddressMystic, Connecticut, USA
Electronic version(s)[pdf]
BibTeX data [bibtex]
IMM Group(s)Intelligent Signal Processing

Back  ::  IMM Publications