Evaluation of Distance Measures Between Gaussian Mixture Models of MFCCs

Jesper Højvang Jensen, Dan P. W. Ellis, Mads Græsbøll Christensen, Søren Holdt Jensen

Publikation: Bidrag til bog/antologi/rapport/konference proceedingKonferenceartikel i proceedingForskningpeer review

41 Citationer (Scopus)
1249 Downloads (Pure)

Abstrakt

In music similarity and in the related task of genre classification, a distance measure between Gaussian mixture models is frequently needed. We present a comparison of the Kullback-Leibler distance, the earth movers distance and the normalized L2 distance for this application. Although the normalized L2 distance was slightly inferior to the Kullback-Leibler distance with respect to classification performance, it has the advantage of obeying the triangle inequality, which allows for efficient searching.
OriginalsprogEngelsk
TitelProceedings of the 8th International Conference on Music Information Retrieval
Antal sider2
ForlagAustrian Computer Society
Publikationsdato2007
Sider107-108
ISBN (Trykt)978-3-85403-218-2
StatusUdgivet - 2007
BegivenhedInternational Conference on Music Information Retrieval - Vienna, Østrig
Varighed: 23 sep. 200727 sep. 2007
Konferencens nummer: 8

Konference

KonferenceInternational Conference on Music Information Retrieval
Nummer8
LandØstrig
ByVienna
Periode23/09/200727/09/2007

Fingeraftryk

Dyk ned i forskningsemnerne om 'Evaluation of Distance Measures Between Gaussian Mixture Models of MFCCs'. Sammen danner de et unikt fingeraftryk.
  • IntelligentSound

    Jensen, S. H., Christensen, M. G., Jensen, J. H. & Laurberg, H.

    STVF

    01/01/200531/12/2008

    Projekter: ProjektForskning

Citationsformater