Title :
Music Genre Classification Based on Entropy and Fractal Lacunarity
Author :
Goulart, Antonio Jose Homsi ; Maciel, Carlos Dias ; Guido, Rodrigo Capobianco ; Paulo, Katia Cristina Silva ; Silva, Ivan Nunes da
Author_Institution :
Sch. of Eng. at Sao Carlos, Univ. of Sao Paulo, Sao Carlos, Brazil
Abstract :
In this letter, we present an automatic music genre classification scheme based on a Gaussian Mixture Model (GMM) classifier. The proposed technique adopts entropies and lacunarities as features for the classifications. Tests were carried out with four styles of Brazilian music, namely Axe, Bossa Nova, Forro, and Samba.
Keywords :
Gaussian processes; audio signal processing; entropy; music; Axe; Bossa Nova; Brazilian music; Forro; Gaussian mixture model classifier; Samba; automatic music genre classification; entropy; fractal lacunarity; Accuracy; Entropy; Feature extraction; Fractals; Humans; Music; Training; GMM; automatic music genre classification; entropy; lacunarity;
Conference_Titel :
Multimedia (ISM), 2011 IEEE International Symposium on
Conference_Location :
Dana Point CA
Print_ISBN :
978-1-4577-2015-4
DOI :
10.1109/ISM.2011.94
