Modelo automático de classificação de gêneros musicais amazônicos

  • Douglas Silva Federal University of Amapá
  • Lucas Zampar Federal University of Amapá
  • Felipe Rodrigues Federal University of Amapá
  • Cláudio Gomes Federal University of Amapá

Abstract


Globalization affects the musical preference of today’s society, which has continued esteemed by international rhythms compared to the national or local. Music is a form of communication used for the construction and organization of the social structure, which It influences lifestyles, tastes, and interpersonal relationships. Therefore, each musical genre manifestation has different functions for a listener, for example, to dance, celebrate, rest, help in solitude, sadness, etc. Several applications use music genre classifiers to indicate, preview, or suggest new music for their listeners. For various reasons, most classifiers do not have information on regional music genres. This work proposes an automatic classification model for Amazonian popular musical genres. Initially, a database was created containing the musical genres: andino, brega, carimb´o, c´umbia, marabaixo, pasillo, salsa e vaqueirada, from the Legal Amazon region of the countries: Brazil, French Guiana, Venezuela, Colombia, Ecuador, Bolivia and Peru. For the construction of the database, several characteristics of each song were extracted from a total of 64 parameters. Machine learning models were analyzed in which XGB, KNN, SVM e MLP obtained an accuracy of 67.62%, 74.12%, 71.35%, 76.13%, respectively.

Keywords: machine learing, music information retrieval, classifiers

References

Susan Schreibman, Ray Siemens, and John Unsworth. A new companion to digital humanities. John Wiley & Sons, 2015.

Rafael José de Menezes Bastos. As contribuições da música popular brasileira às músicas populares do mundo: diálogos transatlânticos Brasil-Europa-África - Segunda parte. Universidade Federal de Santa Catarina, 2007.

Andre Holzapfel, Bob Sturm, and Mark Coeckelbergh. Ethical dimensions of music information retrieval technology. Transactions of the International Society for Music Information Retrieval, 1(1):44–55, 2018.

Shankru Guggari, Vijayakumar Kadappa, V Umadevi, and Ajith Abraham. Music rhythm tree based partitioning approach to decision tree classifier. Journal of King Saud University-Computer and Information Sciences, 2020.

Nagamanoj Karunakaran and Arti Arya. A scalable hybrid classifier for music genre classification using machine learning concepts and spark. In 2018 International Conference on Intelligent Autonomous Systems (ICoIAS), pages 128– 135. IEEE, 2018.

Bob L. Sturm. The GTZAN dataset: Its contents, its faults, their effects on evaluation, and its future use. CoRR, abs/1306.1461, 2013.

Michaël Defferrard, Kirell Benzi, Pierre Vandergheynst, and Xavier Bresson. Fma: A dataset for music analysis. arXiv preprint arXiv:1612.01840, 2016.

Carlos N Silla Jr, Celso AA Kaestner, and Alessandro L Koerich. Automatic music genre classification using ensemble of classifiers. In 2007 IEEE International Conference on Systems, Man and Cybernetics, pages 1687–1692. IEEE, 2007.

Carlos Nascimento Silla Jr, Alessandro L Koerich, and Celso AA Kaestner. The latin music database. In ISMIR, pages 451–456, 2008.

Aurélien Géron. Mãos à Obra: Aprendizado de Máquina com Scikit-Learn & TensorFlow. Alta Books, 2019.
Published
2021-10-24
SILVA, Douglas; ZAMPAR, Lucas; RODRIGUES, Felipe; GOMES, Cláudio. Modelo automático de classificação de gêneros musicais amazônicos. In: BRAZILIAN SYMPOSIA ON COMPUTER MUSIC (SBCM), 18. , 2021, Recife. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2021 . p. 225-228. DOI: https://doi.org/10.5753/sbcm.2021.19453.