H. Sato, Y. Mitsukura, M. Fukumi, and N. Akamatsu (Japan)
new human interface, emotional speech classification, emotional space map, pitch, cepstrum method
The research such that robots are made to understand man’s emotions is one of the very important field of researches in order to achieve a new human interface. In this paper, we propose a new emotional speech classification method and new space map generation method by analyzing feature parameters obtained from the emotional speech and using neural networks. In the present research, the emotion is classified broadly into four patterns such as neutral, anger, sad and joy. In many cases, emotions are found by the change of pitches. The pitch is extracted from each emotional speech by the cepstrum method. In put values of neural networks (NNs) are then emotional pitch patterns, which are time-varying. It is shown that NNs can achieve emotional speech classification and gen eration of space map by learning each emotional pitch pattern by means of computer simulations. Finally, in order to show the effectiveness of the proposed emotional speech classification method, we have done computer simulations. Furthermore, by using simulation results, we generate the emotional space map. Finally, many tests for emotional speech classification are done by using the obtained emotional space map.
Important Links:
Go Back
