Melodic models for polyphonic music classification
Ruben Hillewaere · 2009
The classification of polyphonic music still presents chal- lenges for current music data mining methods. In this paper w ee xplore the performance of classifiers specifically created for melody on the poly- phonic classification task. On a small dataset of string quartet move- ments of Haydn and Mozart, the melodicn-gram model outperforms the melodic global feature model for composer recognition. Furthermore, a simple model that combines the predictions made from di!erent instru- mental parts outperforms models created from any single voice. The results indicate that models taking into account polyphoni ci nformation achieve higher classification accuracy.