Comparative Analysis of Musical Performances by Using Emotion Tracking on the Arousal-Valence Plane

Abstract

More systems searching music on Internet databases are adding emotion selection as an option to basic search parameters. Finding pieces with a similar emotional distribution throughout the same composition is an option that further extends the capabilities of search systems. In this study, we presented a comparative analysis of musical performances by using emotion tracking. A dimensional approach of dynamic music emotion recognition was used in the analysis. Values of arousal and valence, predicted by regressors, were used to compare performances. We analyzed the emotional content of performances of 6 musical works. We found which performances of the same composition are closer to each other and which are quite distant in terms of the shaping of arousal and valence over time. The presented method gives access to knowledge on the shaping of emotions by a performer, which had previously been available only to music professionals.