A comparative study of classifiers for music genre classification based on feature extractors

Abstract

The objective of this paper is to do a comparative study to detect and classify music files automatically based on its genre by using various classification algorithms. Music genre classification is a popular problem in the domain of Music Information Retrieval (MIR) used in many music streaming platforms such as Pandora which is a automated music recommendation service based on the Music Genome Project, that suggests songs to users based on similarity of songs that the user is interested in. In this paper we have done a comparative study using various machine learning classification algorithms to classify music file based on its genre. We have used both Fast Fourier Transform (FFT) and Mel Frequency Cepstral Coefficients (MFCC) to featurize our data, the latter out of which was recommended in a previous study.