Àá½Ã¸¸ ±â´Ù·Á ÁÖ¼¼¿ä. ·ÎµùÁßÀÔ´Ï´Ù.
KMID : 1022420140060020009
Phonetics and Speech Sciences
2014 Volume.6 No. 2 p.9 ~ p.20
Computer-Based Fluency Evaluation of English Speaking Tests for Koreans
Jang Byeong-Yong

Kwon Oh-Wook
Abstract
In this paper, we propose an automatic fluency evaluation algorithm for English speaking tests. In the proposed algorithm,acoustic features are extracted from an input spoken utterance and then fluency score is computed by using support vectorregression (SVR). We estimate the parameters of feature modeling and SVR using the speech signals and the correspondingscores by human raters. From the correlation analysis results, it is shown that speech rate, articulation rate, and mean lengthof runs are best for fluency evaluation. Experimental results show that the correlation between the human score and the SVRscore is 0.87 for 3 speaking tests, which suggests the possibility of the proposed algorithm as a secondary fluency evaluationtool.
KEYWORD
speaking fluency, pronunciation, SVR, regression
FullTexts / Linksout information
Listed journal information
ÇмúÁøÈïÀç´Ü(KCI)