Recognizing Brazilian music genres with supervised machine learning techniques

• Julia Luiza da Conceição UFAM
• Rosiane de Freitas UFAM
• Bruno Gadelha UFAM
• João Gustavo Kienen UFAM
• Sérgio Anders UFAM
• Brendo Cavalcante UFAM

DOI: https://doi.org/10.5753/semish.2020.11328

Abstract

In this work, an initial study on the automatic recognition of the main Brazilian musical genres is presented: Axe, Forró, MPB, Rock, Samba, and Sertanejo. Through the extraction of representative musical characteristics, automatic classification experiments were performed using the Weka tool and classical supervised learning algorithms. An analysis of the main available databases was also carried out: GTZAN, FMA, AudioSet, RWC, ISMIR, Magnatune, and LMD. There is a scarcity of cultural diversity on these bases, most of which concentrate globally more popular styles such as Pop and Rock, reinforcing the need to include more diverse and culturally identifiable genres, such as Brazilians.

Keywords: automatic recognition, Brazilian musical genres, supervised learning

References

Aucouturier, J.-J. and Pachet, F. (2003). Representing musical genre: A state of the art. Journal of New Music Research, 32(1):83–93.

Biau, G. (2012). Analysis of a random forests model. Journal of Machine Learning Research, 13(Apr):1063–1095.

BREsCIA, V. P. (2009). A música como recurso terapêutico. In Encontro Paranaense Congresso Brasileiro de Psicoterapias Corporais, XIV, IX.

Haggblade, M., Hong, Y., and Kao, K. (2011). Music genre classification. Department of Computer Science, Stanford University.

Hall, M., Frank, E., Holmes, G., Pfahringer, B., Reutemann, P., and Witten, I. H. (2009). The weka data mining software: An update. SIGKDD Explor. Newsl., 11(1):10–18.

Jeong, I.-Y. and Lee, K. (2016). Learning temporal features using a deep neural network and its application to music genre classification. In Ismir, pages 434–440.

Lima, M., Valle, V., Costa, E., Lira, F., and Gadelha, B. (2019). Software engineering repositories: Expanding the promise database. In Proceedings of the XXXIII Brazilian Symposium on Software Engineering, pages 427–436.

McFee, B., Raffel, C., Liang, D., Ellis, D. P., McVicar, M., Battenberg, E., and Nieto, O. (2015). librosa: Audio and music signal analysis in python. In Proceedings of the 14th python in science conference, volume 8.

Moreira, S. V., Franc ̧a, C. C., Moreira, M. A., and Lana-Peixoto, M. A. (2009). Identidade musical em pacientes com esclerose múltipla. Arquivos de Neuro-Psiquiatria, 67(1):46–49.

Naik, A. and Samant, L. (2016). Correlation review of classification algorithm using data mining tool: Weka, rapidminer, tanagra, orange and knime. Procedia Computer Science, 85:662–668.

Rajvanshi, N. and Chowdhary, K. (2017). Comparison of svm and na ̈ıve bayes text clas- sification algorithms using weka. International Journal of Engineering Research and, 6:09.

Ramírez, J. and Flores, M. J. (2019). Machine learning for music genre: multifaceted review and experimentation with audioset. Journal of Intelligent Information Systems, pages 1–31.

Silla, C. N., Koerich, A. L., and Kaestner, C. A. (2008). A machine learning approach to automatic music genre classification. Journal of the Brazilian Computer Society, 14(3):7–18.

Silla Jr, C. N., Koerich, A. L., and Kaestner, C. A. (2008). The latin music database. In ISMIR, pages 451–456.

Sturm, B. L. (2012). A survey of evaluation in music genre recognition. In International Workshop on Adaptive Multimedia Retrieval, pages 29–66. Springer.

Sturm, B. L. (2013). The gtzan dataset: Its contents, its faults, their effects on evaluation, and its future use. arXiv preprint arXiv:1306.1461.