Projekter pr. år
Abstract
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.
Originalsprog | Engelsk |
---|---|
Titel | Proceedings of the 8th International Conference on Music Information Retrieval |
Antal sider | 2 |
Forlag | Austrian Computer Society |
Publikationsdato | 2007 |
Sider | 107-108 |
ISBN (Trykt) | 978-3-85403-218-2 |
Status | Udgivet - 2007 |
Begivenhed | International Conference on Music Information Retrieval - Vienna, Østrig Varighed: 23 sep. 2007 → 27 sep. 2007 Konferencens nummer: 8 |
Konference
Konference | International Conference on Music Information Retrieval |
---|---|
Nummer | 8 |
Land/Område | Østrig |
By | Vienna |
Periode | 23/09/2007 → 27/09/2007 |
Fingeraftryk
Dyk ned i forskningsemnerne om 'Evaluation of Distance Measures Between Gaussian Mixture Models of MFCCs'. Sammen danner de et unikt fingeraftryk.Projekter
- 1 Afsluttet
-
IntelligentSound
Jensen, S. H., Christensen, M. G., Jensen, J. H. & Laurberg, H.
01/01/2005 → 31/12/2008
Projekter: Projekt › Forskning