Selected methods of parametrization in problem of automatic classification classical music from the Renaissance era against the classical works from other eras

Abstract

In this article we present the results of our work in the field of automatic classification of classical music pieces. The studied works were compositions of classical music composed in four eras: Renaissance, Baroque, Classicism and Romanticism. In the work we described selected methods of parameterization of music files, so that they emphasize the characteristic of Renaissance. The parameters we use are of a horizontal nature, i.e. they do not penetrate the vertical structure of the piece (e.g. chords progression). We used a base of 571 works of classical music, both secular and religious. The files were stored in MusicXML format and contained 187 Renaissance pieces, 146 Baroque, 119 classics and 119 stylistically belonging to the Romantic era, respectively. The results of the studies were presented using 4, 13 and 113 parameters. An artificial neural network and Support Vector Machine were used to classify the era to which the song belongs.