Recognition Across Malay and Japanese Spoken Languages

Only available on StudyMode
  • Topic: Spectrum, Paul Ekman, Emotions
  • Pages : 2 (293 words )
  • Download(s) : 500
  • Published : November 11, 2010
Open Document
Text Preview
平成21年度 特別実験報告書概要

|課程,学籍番号,氏名 |課程:情報工学課程, 学籍番号:083447,氏名:LIYANA SAFRA BINTI ZAABAR | |工学分野名:情報処理工学分野 |指導教員名:青野雅樹 | |題 目 和   マレー語と日本語における音声からの感情認識 | |(英 Emotion Recognition across Malay and Japanese Spoken Languages) | |Abstract | |In this study, we report our research on emotion recognition in Malay and Japanese spoken languages. The goal of this study is to attempt to find | |out which features of speech signal could be useful for emotion recognition in both languages. We consider four different types of features for | |speech signal: pitch, formant, intensity, and LTAS (Long Term Average Spectrum), where LTAS is given by the logarithmic power spectral density as a | |function of frequency. | |Based on the above features, we conducted experiments using recorded voices with six different emotions (i.e. joy, sadness, fear, anger, disgust, | |and surprise) from both Malay and Japanese languages, and asked sixteen subjects to estimate which emotions they might feel after listening to each | |recorded voice. Then we compared the results with those obtained by computer estimates. We have found interesting similarities and discrepancies | |between the comparison of computer and human estimates of emotions, as well as different kind of similarities and discrepancies between Malay and | |Japanese languages....
tracking img