Static Music Emotion Recognition Using Recurrent Neural Networks

The article presents experiments using recurrent neural networks for emotion detection for musical segments using Russell's circumplex model. A process of feature extraction and creating sequential data for learning networks with long short-term memory (LSTM) units is presented. Models were implemented using the WekaDeeplearning4j package and a number of experiments were carried out with data with different sets of features and varying segmentation. The usefulness of dividing data into sequences as well as the sense of using recurrent networks to recognize emotions in music, whose results have even exceeded the SVM algorithm for regression, were demonstrated. The author analyzed the effect of the network structure and the set of used features on the results of regressors recognizing values on two axes of the emotion model: arousal and valence.