Pachet, F. and Roy, P. Hit Song Science is Not Yet a Science. Proceedings of Ismir 2008, pages 355-360, Philadelphia, USA, 2008

Sony CSL authors: François Pachet, Pierre Roy

Abstract

We describe a large-scale experiment aiming at validating the hypothesis that the popularity of music titles can be predicted from global acoustic or human features. We use a 32.000 title database with 632 manually-entered labels per title including 3 related to the popularity of the title. Our experiment uses two audio feature sets, as well as the set of all the manually-entered labels but the popularity ones. The experiment shows that some subjective labels may indeed be reasonably well-learned by these techniques, but not popularity. This contradicts recent and sustained claims made in the MIR community and in the media about the existence of “Hit Song Science”.

Keywords: feature extraction, timbre, Hit Song Science, audio descriptors

Downloads

[PDF] Adobe Acrobat PDF file

BibTeX entry

@INPROCEEDINGS { pachet:08c, ADDRESS="Philadelphia, USA", AUTHOR="Pachet, F. and Roy, P.", BOOKTITLE="Proceedings of Ismir 2008", PAGES="355-360", TITLE="Hit Song Science is Not Yet a Science", YEAR="2008", }