Music generation based on emotional EEG

Abstract

Transforming electroencephalogram (EEG) into music has been playing an important role in social life. How to generate music that can express a certain emotion state is a challenge for most of the existing generative models in the studies. To address the problem, a music generation method based on emotional EEG is proposed in this paper. In this method, sequence to sequence long-short term memory is utilized to train the emotional music to obtain emotional music generators, and support vector machine is used to get emotional information. The features related to emotion are extracted to map into musical parameters and emotion music generator is used to generate the emotional EEG music. The experimental results show that the music generated by the proposed method achieves a high performance with respect to both emotion-expressing and musicality.