The results showed that the recognition scores are rather high with the case for which there is a full combination of parameters as MFCC and its first and second derivatives, fundamental frequency, energy, formants and its correspondent bandwidths, spectral characteristics and F0 variants. | Journal of Computer Science and Cybernetics, , (2017), 229–246 DOI GMM FOR EMOTION RECOGNITION OF VIETNAMESE DAO THI LE THUY1,2 , TRINH VAN LOAN2 , NGUYEN HONG QUANG2 1 Faculty of Information Technology, Ha Noi Vocational College of High Technology 2 Ha Noi University of Science and Technology; thuydt@ Abstract. This paper presents the results of GMM-based recognition for four basic emotions of Vietnamese such as neutral, sadness, anger and happiness. The characteristic parameters of these emotions are extracted from speech signals and divided into different parameter sets for experiments. The experiments are carried out according to speaker-dependent or speaker-independent and contentdependent or content-independent recognitions. The results showed that the recognition scores are rather high with the case for which there is a full combination of parameters as MFCC and its first and second derivatives, fundamental frequency, energy, formants and its correspondent bandwidths, spectral characteristics and F 0 variants. In average, the speaker-dependent and content-dependent recognition scrore is . Next, the average score is for the speaker-dependent and contentindependent recognition. For the speaker-independent and content-dependent recognition, the average score is . The average score is for speaker-independent and content-independent recognition. Information on F 0 has significantly increased the score of recognition. Keywords. GMM, recognition, emotion, Vietnamese, corpus, F 0. 1. INTRODUCTION Recognition of emotional speech has been of interest to researchers because it is particularly useful for applications that require a natural interaction between man and machine. There are many studies on recognition of emotional speech available in a number of different languages around the world such as English, German, Chinese, French, Spanish,. . . [1]. The majority of these studies use .
