[Back]

R. Neumayer:
"Musical genre classification using a multi layer perceptron";
in: "Proceedings of the Fifth Workshop on Data Analysis", Elfa Academic Press, 2004, ISBN: 80-89066-87-9, 51 - 66.

Abstract. Musical genres are labels used to distinguish between different types or categories of musical style. The growing amount of music that is available electronically, particularly in the World Wide Web, creates a rising need for automated classification. A typical application could be plugins for music management software or some sort of tool that automatically tags audio files according to its genre. This article covers the possibilities of musical genre classification and its application to mp3-files. Mainly it deals with the classification accuracies obtained by the use of different music collections using a multi layer perceptron. Moreover it discusses the similarities or differences between different genres according to a specific set of features.

Machine Learning, Genre Classification, Music Information Retrieval