PROJECT TITLE :

Hierarchical Dirichlet Process Mixture Model for Music Emotion Recognition

ABSTRACT:

This study proposes a unique multi-label music emotion recognition (MER) system. An emotion can't be outlined clearly in the important world as a result of the categories of emotions are sometimes thought-about overlapping. Accordingly, this study proposes an MER system that is based mostly on hierarchical Dirichlet process mixture model (HPDMM), whose components will be shared between models of each emotion. Moreover, the HDPMM is improved by adding a discriminant factor to the proposed system based on the concept of linear discriminant analysis. The proposed system represents an emotion using weighting coefficients that are related to a global set of elements. Moreover, three strategies are proposed to compute the weighting coefficients of testing data, and also the weighting coefficients are used to see whether or not or not the testing information contain certain emotional content. Within the tasks of music emotion annotation and retrieval, experimental results show that the proposed MER system outperforms state-of-the-art systems in terms of F-score and mean average precision.