A Chinese-English Interpretation Testing Autoscoring System Based on Semantic Analysis

DOI

10.2991/icmii-15.2015.86How to use a DOI?

Keywords

Speech Recognition, Semantic Analysis, Speech Autoscoring, Wordnet

Abstract

This paper aims to find out an approach to autoscoring the Chinese-English interpretation testing speech, which is one of the important research areas, meets the demand of computer-assistant English testing and adds value to machine intelligence. We proposed a method to evaluate the semantic similarity based on Wordnet combining words and sentences, which was applied to recognizing the keywords and its synonym in testing speech via HMM model and calculating the score of semantic. In addition, speech fluency was also considered in our developed system and has a contribution to the final score. This paper collected 300 testing speech data from a course final examination in Guangdong University of Foreign Studies to examine the autoscoring system. Result has shown well performance of the average rate of speech recognition and agreement of system score and human score.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Xinguang Li
AU  - Jiyou Xu
AU  - Shengbin Zhang
AU  - Zhiming He
PY  - 2015/10
DA  - 2015/10
TI  - A Chinese-English Interpretation Testing Autoscoring System Based on Semantic Analysis
BT  - Proceedings of the 3rd International Conference on Mechatronics and Industrial Informatics
PB  - Atlantis Press
SP  - 502
EP  - 508
SN  - 2352-538X
UR  - https://doi.org/10.2991/icmii-15.2015.86
DO  - 10.2991/icmii-15.2015.86
ID  - Li2015/10
ER  -