EMOTION INFERENCE METHOD BASED ON WORDS MEANING AND UTTERANCE FEATURES

摘要：

Emotion recognition is necessary to realize smooth communication between human and computer. This study proposes an emotion prediction method based on two communication measures: words and utterances. Concretely, we investigate the emotions caused by meaningful expressions such as “arigato (“Thank you.) and “baka (“Stupid.). Features of utterance such as utterance speed, maximum value of power and the highest basic frequency are extracted from four speeches. In the case of male utterance, we learned that invoking emotions of the listener was defined by the only utterance speed regaredless of word’s meaning. But, in the case of female utterance, we learned that invoking emotions of the listener was defined by the utterance speed and maximum of power regaredless of word’s meaning.

关键词： affective computing words meaning utterance features

作者: Masashi ADACHI Seiji TSUCHIYA Fuji REN

作者单位: Graduate School of Advanced Technology and Science,The University of Tokushima Minami josanjima,Toku Institute of Technology and Science,The University of Tokushima Institute of Technology and Science,The University of Tokushima;School of Information Engineering,Be

会议类型: 国际会议

会议名称: 2008高等智能国际会议(2008 International Conference on Advanced Intelligence)

会议地点: 北京

会议语种:英文

在线出版日期: 2008-10-18（万方平台首次上网日期，不代表论文的发表时间）