Music Genre Classification by Textures in the Frequency Space

• Yandre M. G. Costa UEM / UFPR
• Luiz S. Oliveira UFPR
• Alessandro L. Koerich PUC-PR
• Fabien Gouyon INESC-Porto

Abstract

This paper describes an attempt to perform automatic musical genre classification based on spectrograms extracted from segments of digital music pieces, taken from the “Latin Music Database”. Feature vectors with textural characteristics were extracted from digital images of spectrograms by using gray level co-occurrence matrix. The recognition rate obtained with a Support Vector Machine classifier was 60,11%. This rate is slightly higher than other obtained with different approaches recently performed over the same database. In addition, the classifier proposed here was combined with another classifier. The results obtained show an recognition rate about 66,11% and the upper limit found combining this two classifiers is about 75%.

References

Chang, C.-C. and Lin, C.-J. (2001). LIBSVM: a library for support vector machines. Software available at [link].

Costa, C., Valle Jr, J., and Koerich, A. (2004). Automatic classification of audio data. In IEEE International Conference on Systems, Man, and Cybernetics, pages 562–567.

French, M. and Handy, R. (2007). Spectrograms: turning signals into pictures. Journal of engineering technology, 24(1):p. 32–35.

Gantz, J., Chute, C., Manfrediz, A., Minton, S., Reinsel, D., Schlichting, W., and Toncheva, A. (2008). The diverse and exploding digital universe: An updated forecast of worldwide information growth through 2011. IDC white paper, sponsored by EMC.

Haralick, R., Shanmugam, K., and Dinstein, I. (1973). Textural features for image classification. IEEE Transactions on systems, man and cybernetics, 3(6):p. 610–621.

Kittler, J., Hatef, M., Duin, R., and Matas, J. (2002). On combining classifiers. Pattern Analysis and Machine Intelligence, IEEE Transactions on, 20(3):p. 226–239.

Koerich, A. and Poitevin, C. (2005). Combination of Homogeneous Classifiers for Musical Genre Classification. IEEE International Conference on Systems, Man, and Cybernetics, pages 554–559.

Li, T., Ogihara, M., and Li, Q. (2003). A comparative study on content-based music genre classification. In Proceedings of the 26th annual international ACM SIGIR conference on Research and development in informaion retrieval, pages 282–289. ACM.

Lopes, M., Gouyon, F., Koerich, A. L., and Oliveira, L. E. S. (2010). Selection of training instances for music genre classification. ICPR 2010 - 20th International Conference on Pattern Recognition.

Lucena, C., Medeiros, C., Lucchesi, C., Maldonado, J., Almeida, V., and outros (2006). Grandes Desafios da Pesquisa em Computação no Brasil - 2006-2016. Relatório sobre o seminário Grandes Desafios da Pesquisa em Computação. São Paulo. SBC/CAPES/FAPESP.

Pampalk, E., Flexer, A., and Widmer, G. (2005). Improvements of audio-based music similarity and genre classification. In proc. ISMIR, volume 5.

Silla Jr, C., Koerich, A., and Kaestner, C. (2008). The latin music database. In Proceedings of the 9th International Conference on Music Information Retrieval, pages 451–456.

Tzanetakis, G. and Cook, P. (2002). Musical genre classification of audio signals. IEEE Transactions on speech and audio processing, 10(5):p. 293–302.

Vapnik, V. (2000). The nature of statistical learning theory. Springer Verlag.

Yu, G. and Slotine, J. (2009). Audio classification from time-frequency texture. ICASSP 2009 - International Conference on Acoustics, Speech and Signal Processing, pages 1677–1680.