Presentation 2001/6/22
Classification of Emotional Speech by Using Neural Networks
Hideaki Sato, Norio Akamatsu,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) It is necessary to approach the analytic processing of kansei information in order to achieve a new Human Interface that digital computers can deal with the kansei information. In this paper, we propose a new classification method of emotional speech by analyzing characteristic parameters obtained from emotional speech and by learning those by neural networks, which is a kansei information processing. In this research, the emotion is classified broadly into four patterns such as neutral, anger, sad and joy. The pitch as characteristic parameters isextracted from each emotional speech by the cepstrum method. In addition, input values of neural networks(NN) are emotional pitch patterns, which are time-varying. NN can achieve classification of emotion by learning each emotional pitch pattern. Finally, in order to demonstrate the effectiveness of the proposed scheme, computer simulations were performed.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) kansei information / emotional speech / pitch / cepstrum method
Paper # NC2001-34
Date of Issue

Conference Information
Committee NC
Conference Date 2001/6/22(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Neurocomputing (NC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Classification of Emotional Speech by Using Neural Networks
Sub Title (in English)
Keyword(1) kansei information
Keyword(2) emotional speech
Keyword(3) pitch
Keyword(4) cepstrum method
1st Author's Name Hideaki Sato
1st Author's Affiliation Faculty of Engineering, University of Tokushima()
2nd Author's Name Norio Akamatsu
2nd Author's Affiliation Faculty of Engineering, University of Tokushima
Date 2001/6/22
Paper # NC2001-34
Volume (vol) vol.101
Number (no) 154
Page pp.pp.-
#Pages 6
Date of Issue