Desktop Site

News & Events

Awards

Mr.Xingfeng LI of the School of Infomation Science received Best Student Oral Paper Award in 18th Oriental Chapter of COCOSDA.

Mr.Xingfeng LI, a 2nd year master's student in Akagi Lab. of the School of Information Science, received Best Student Oral Paper Award in 18th Oriental Chapter of COCOSDA(International Committee for the Co-ordination and Standardization of Speech Databases and Assessment Techniques 2015) .

Reference:http://www.ococosda2015.org/Default.aspx

Reference:http://www.ococosda2015.org/content.aspx?info_lb=31&flag=3

[Date awarded] October 30, 2015
[Title] "Toward Improving Estimation Accuracy of Emotion Dimensions in Bilingual Scenario Based on Three-Layered Model"
[Abstract] This paper proposes a newly revised three-layered model to improve emotion dimensions (valence, activation) estimation for bilingual scenario, using knowledge of commonalities and differences of human perception among multiple languages. Most of previous systems on speech emotion recognition only worked in each mono-language. However, to construct a generalized emotion recognition system which be able to detect emotions for multiple languages, acoustic features selection and feature normalization among languages remained a topic. In this study, correlated features with emotion dimensions are selected to construct proposed model. To imitate emotion perception across languages, a novel normalization method is addressed by extracting direction and distance from neutral to other emotion in emotion dimensional space. Results show that the proposed system yields mean absolute error reduction rate of 46% and 34% for Japanese and German language respectively over previous system. The proposed system attains estimation performance more comparable to human evaluation on bilingual case.
[Comment] It is so honored and grateful to be award as the best student oral paper in the 18th Oriental COCOSDA. I would like to thanks to COCOSDA organizers who gave me a wonderful and meaningful experience to attend this interesting conference. And, especially, it should be greatly appreciated to my supervisor, Prof. Masato Akagi for providing me precious guidance and kindly help all the time. Of course, this honor should be shared with all my lab members, it comes from joint efforts. At last, I would like to express my gratitude to JAIST. This is a completely nice institute for researching. Research is a long story, and I will try to work harder to obtain further results.

award_akagilab.jpg

January 15, 2016

PAGETOP