Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

A feature survey for emotion classification of western popular music

: Beveridge, Scott; Knox, Don

Queen Mary University of London -QMUL-, Centre for Digital Music; Centre National de la Recherche Scientifique -CNRS-, Laboratoire de Mécanique et D'Acoustique -LMA-, Marseille:
Music and emotions. Proceedings : 9th International Symposium on Computer Music Modeling and Retrieval, CMMR 2012. 19-22 June, Queen Mary University of London, London, UK
London, 2012
International Symposium on Computer Music Modeling and Retrieval (CMMR) <9, 2012, London>
Conference Paper
Fraunhofer IDMT ()
emotion recognition; music classification

In this paper we propose a feature set for emotion classification of Western popular music. We show that by surveying a range of common feature extraction methods, a set of five features can model emotion with good accuracy. To evaluate the system we implement an independent feature evaluation paradigm aimed at testing the property of generalizability; the ability of a machine learning algorithm to maintain good performance over different data sets.