Recognition and analysis of emotion transition in mandarin speech signal

Language is one and probably most important means for communication between human beings. The human emotion usually accompanies with speech. So human speech identification is not only involving language syntax and meaning but also the emotion at that moment. In this paper, we study the long sentence speech corpus emotion recognition using various speech segmentation approaches, and try to detect the emotion transition point from the continuous speech. In the experiments, we apply three different speech segmentation methods to the continuous Mandarin emotional speech. These methods include uniform, endpoint detection, and whole sentence segmentation. There are five emotions being investigated, including anger, happiness, sadness, boredom, and neutral. We then employ two classification algorithms, the conventional K-nearest neighbor (KNN) and weighted discrete-KNN (WD-KNN), in the recognition phase. From to the experimental results, we find that the WD-KNN yields an average recognition rate of 73% for the used testing sentences.