Presentation 2014-09-27
Development of automatic system to detect mutual assessment in learning activities from relationship speech and head motion
Shu Akabene, Jun Yamashita, Hideyuki Suzuki, Hiroshi Kato, Hideaki Kuzuoka,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) The aim of this research is to raise recognition accuracy of speech recognition using relationship speech and head motion in learning activities. In this paper, we propose method combining "cmscore" which Julius that is an open source real-time recognition engine outputs to show word confidence and Bayesian estimation for raising accuracy of speech recognition. From this method, we show possible that it may improve recognition accuracy of speech recognition and be able to recognize paralanguage that is not transmitted in its meaning only characters.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) mutual assessment in learning activities / speech recognition / Head motion detection / Bayesian estimation
Paper # ET2014-33
Date of Issue

Conference Information
Committee ET
Conference Date 2014/9/20(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 Educational Technology (ET)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Development of automatic system to detect mutual assessment in learning activities from relationship speech and head motion
Sub Title (in English)
Keyword(1) mutual assessment in learning activities
Keyword(2) speech recognition
Keyword(3) Head motion detection
Keyword(4) Bayesian estimation
1st Author's Name Shu Akabene
1st Author's Affiliation University of Tsukuba, Graduate School of Systems and Information()
2nd Author's Name Jun Yamashita
2nd Author's Affiliation University of Tsukuba, Department of Intelligent Interaction Technologies
3rd Author's Name Hideyuki Suzuki
3rd Author's Affiliation Ibaraki University
4th Author's Name Hiroshi Kato
4th Author's Affiliation The Open University of Japan
5th Author's Name Hideaki Kuzuoka
5th Author's Affiliation University of Tsukuba, Department of Intelligent Interaction Technologies
Date 2014-09-27
Paper # ET2014-33
Volume (vol) vol.114
Number (no) 228
Page pp.pp.-
#Pages 6
Date of Issue