한국인 영어 학습자의 영어 발음평가 DB 구축을 위한 한국인 평가자 연구
A Study on Usability of Korean Evaluator in Developing a Database for the Assesment of English Pronunciation Spoken by Korean Learners of English
0
조회
0
다운로드
7
인용
0
참고문헌
키워드
초록
In this paper, we investigated the difference between Korean and English evaluators in marking the pronunciation of the non-native speech spoken by Korean learners of English. In order to develop an automatic system measuring pronunciation accuracy of non-native speech, it is indispensible to have non-native speech as well as native speech databases which the system is trained and tested on. The performance of the system should be checked as to whether the output of the system is similar to human judgement on the pronunciation of the non-native speech. Human evaluation on the pronunciation of some portion of the non-native speech database has been done. 8 human evaluators participated. 5 of them were Phonetics major Korean speakers and 3 of them were English native speakers. They are all teaching English or related subjects to university level students. In order to see the difference between the Korean and the English evaluators on pronunciation of the non-native speech, we did statistical tests and found that there is no significant difference and that the correlations between any two evaluators, Korean or English are high and similar. We concluded that evaluation by the native Korean is as useful as the evaluation by the native English speakers and that both can be used in testing the performance of the automatic system measuring the accuracy of non-native speech spoken by Korean learners of English.
Abstract
In this paper, we investigated the difference between Korean and English evaluators in marking the pronunciation of the non-native speech spoken by Korean learners of English. In order to develop an automatic system measuring pronunciation accuracy of non-native speech, it is indispensible to have non-native speech as well as native speech databases which the system is trained and tested on. The performance of the system should be checked as to whether the output of the system is similar to human judgement on the pronunciation of the non-native speech. Human evaluation on the pronunciation of some portion of the non-native speech database has been done. 8 human evaluators participated. 5 of them were Phonetics major Korean speakers and 3 of them were English native speakers. They are all teaching English or related subjects to university level students. In order to see the difference between the Korean and the English evaluators on pronunciation of the non-native speech, we did statistical tests and found that there is no significant difference and that the correlations between any two evaluators, Korean or English are high and similar. We concluded that evaluation by the native Korean is as useful as the evaluation by the native English speakers and that both can be used in testing the performance of the automatic system measuring the accuracy of non-native speech spoken by Korean learners of English.
참고문헌 (16)
- 구희산, "지역 방언 화자에 따른 영어 모음의 발음 연구", 음성과학, 8(4), pp.193~206 (2001)
- 이보림, 이숙향, "한국어 화자의 영어유음 지각 및 산출에 관한 연구", 말소리, 0(52), pp.61~84 (2004), 대한음성학회
- 이석재, "한국인의 영어 음성 코퍼스 설계 및 구축", 말소리, 46(46), pp.159~174 (2003), 대한음성학회
- 임창근, "컴퓨터 음성인식(ASR) 기술을 활용한 영어 말하기 교육", Multimedia Assisted Language Learning, 4(2), pp.187~209 (2001)
- 정현성, "한국인 영어 학습자의 발음 정확성 자동 측정방법에 대한 연구", 언어와언어학, (42), pp.165~196 (2008), 언어연구소
- Franco, H., "Automatic pronunciation scoring for language instruction", pp.1471~1474, Proceedings of ICASSP
- Jong-miKim, "Annotation of a Non-native English Speech Database by Korean Speakers", 음성과학, 9(1), pp.111~135 (2002), 한국음성과학회
- Kim,M-S., "On the Effective Use of ASR Programs in a College Class of English Phonetics and Phonology", English Language and Linguistics, 22, pp.115~132 (2006)
- ShinsookLee, "Acquisition of English Consonant Clusters Among Korean EFL Learners", 언어, 27(3), pp.439~472 (2002), 한국언어학회(대표:홍재성)
- Lee,O-H., "Syllable-timing of Korean English. Proceedings of the 16th Conference on Speech Sciences", pp.295~305 (2005), Gyeongsan
- Neri, A., "Automatic speech recognition for second language learning: How and why it actually works. 15th Proceedings of ICPhS", pp.1157~1160, Barcelona
- Neumeyer, L., "Automatic scoring of pronunciation quality", Speech Communication, 30, pp.83~93 (2000)
- Neumeyer, L., "Automatic text-independent pronunciation scoring of foreign language student speech. Proceedings of ICSLP 96", pp.1457~1460, Philadelphia
- Rabiner, L. R., "Fundamentals of Speech Recognition" (1993), Prentice-Hall
- Um, H-Y., "The English intonation of native speakers and Korean learners", Speech Sciences, 11(1), pp.117~151 (2004)
- Yang,B., "An Acoustical study of English word stress produced by Americans and Koreans", Speech Sciences, 9(1), pp.77~88 (2002)