Emotion Recognition From Audio Signals Using SVM and Naïve Bayes

Speech and music are two types of vocal communications that are mostly related to each other. While significant progress has been made in both speech and music emotion recognition, few works have concentrated on building a shared emotion recognition model for both speech and song. In this paper, we have proposed three shared emotion recognition models for speech and song: a simple model comprising of feature extraction and classification using SVM and naïve bayes. We have used RAVDESS dataset for audio emotion signals of both speech and songs for female and male actors. After features extraction we have evaluated the performance using SVM and Naïve Bayes classification. Our results show that the multi-task model classifies emotion more accurately when the same set of features is used, however spoken and sung emotion recognition tasks are different, and can be considered together, they are related.