Utilize este identificador para referenciar este registo: http://hdl.handle.net/10400.21/2238
Título: Short-term Feature Space and Music Genre Classification
Autor: Marques, Gonçalo
Langlois, Thibault
Gouyon, Fabien
Lopes, Miguel
Sordo, Mohamed
Palavras-chave: Short-term Feature Space
Music Genre Classification
Data: 2011
Editora: Routledge Journals
Citação: MARQUES, Gonçalo; LANGLOIS, Thibault; GOUYON, Fabien; LOPES, Miguel; SORDO, Mohamed - Short-term Feature Space and Music Genre Classification. Journal of New Music Research. ISSN 0929-8215. Vol. 40, n.º 2 (2011) p. 127-137.
Resumo: In music genre classification, most approaches rely on statistical characteristics of low-level features computed on short audio frames. In these methods, it is implicitly considered that frames carry equally relevant information loads and that either individual frames, or distributions thereof, somehow capture the specificities of each genre. In this paper we study the representation space defined by short-term audio features with respect to class boundaries, and compare different processing techniques to partition this space. These partitions are evaluated in terms of accuracy on two genre classification tasks, with several types of classifiers. Experiments show that a randomized and unsupervised partition of the space, used in conjunction with a Markov Model classifier lead to accuracies comparable to the state of the art. We also show that unsupervised partitions of the space tend to create less hubs.
Peer review: yes
URI: http://hdl.handle.net/10400.21/2238
ISSN: 0929-8215
Aparece nas colecções:ISEL - Eng. Elect. Tel. Comp. - Artigos

Ficheiros deste registo:
Ficheiro Descrição TamanhoFormato 
Short-term Feature Space and Music Genre Classification.rep.pdf275,65 kBAdobe PDFVer/Abrir

FacebookTwitterDeliciousLinkedInDiggGoogle BookmarksMySpace
Formato BibTex MendeleyEndnote Degois 

Todos os registos no repositório estão protegidos por leis de copyright, com todos os direitos reservados.