Music Emotion Maps in the Arousal-Valence Space

Abstract

This chapter presents the process of building emotion maps of musical compositions. In our approach, emotion recognition was treated as a regression problem and a two-dimensional valence-arousal model was used to measure emotions. Conducting experiments required the construction of regressors, attribute selection, and analysis of selected musical compositions. We also examined the influence of different feature sets—low-level, rhythm, tonal, and their combination—on arousal and valence prediction. The use of a combination of different types of features significantly improves the results compared with using just one group of features. We found and presented features particularly dedicated to the detection of arousal and valence separately, as well as features useful in both cases. The obtained emotion maps provide new knowledge about the distribution of emotions in an examined audio recording. They reveal new knowledge that had only been available to music experts until this point. We propose the features for analyzing and comparing changes in arousal and valence over time and use them to compare selected well-known Ludwig van Beethoven’s Sonatas with several of the most famous songs by The Beatles.