Efficient pronunciation assessment of taiwanese-accented english based on unsupervised model adaptation and dynamic sentence selection

Chung-Hsien Wu, Hung Yu Su, Chao Hong Liu

研究成果: Chapter

摘要

This chapter presents an efficient approach to personalized pronunciation assessment of Taiwaneseaccented English. The main goal of this study is to detect frequently occurring mispronunciation patterns of Taiwanese-accented English instead of scoring English pronunciations directly. The proposed assessment help quickly discover personalized mispronunciations of a student, thus English teachers can spend more time on teaching or rectifying students' pronunciations. In this approach, an unsupervised model adaptation method is performed on the universal acoustic models to recognize the speech of a specific speaker with mispronunciations and Taiwanese accent. A dynamic sentence selection algorithm, considering the mutual information of the related mispronunciations, is proposed to choose a sentence containing the most undetected mispronunciations in order to quickly extract personalized mispronunciations. The experimental results show that the proposed unsupervised adaptation approach obtains an accuracy improvement of about 2.1% on the recognition of Taiwanese-accented English speech.

原文English
主出版物標題Multidisciplinary Computational Intelligence Techniques
主出版物子標題Applications in Business, Engineering, and Medicine
發行者IGI Global
頁面12-30
頁數19
ISBN(列印)9781466618305
DOIs
出版狀態Published - 2012 十二月 1

指紋

Students
Teaching
Acoustics

All Science Journal Classification (ASJC) codes

  • Computer Science(all)

引用此文

Wu, C-H., Su, H. Y., & Liu, C. H. (2012). Efficient pronunciation assessment of taiwanese-accented english based on unsupervised model adaptation and dynamic sentence selection. 於 Multidisciplinary Computational Intelligence Techniques: Applications in Business, Engineering, and Medicine (頁 12-30). IGI Global. https://doi.org/10.4018/978-1-4666-1830-5.ch002
Wu, Chung-Hsien ; Su, Hung Yu ; Liu, Chao Hong. / Efficient pronunciation assessment of taiwanese-accented english based on unsupervised model adaptation and dynamic sentence selection. Multidisciplinary Computational Intelligence Techniques: Applications in Business, Engineering, and Medicine. IGI Global, 2012. 頁 12-30
@inbook{547c0e5d206243fc92c2a110f1f485b3,
title = "Efficient pronunciation assessment of taiwanese-accented english based on unsupervised model adaptation and dynamic sentence selection",
abstract = "This chapter presents an efficient approach to personalized pronunciation assessment of Taiwaneseaccented English. The main goal of this study is to detect frequently occurring mispronunciation patterns of Taiwanese-accented English instead of scoring English pronunciations directly. The proposed assessment help quickly discover personalized mispronunciations of a student, thus English teachers can spend more time on teaching or rectifying students' pronunciations. In this approach, an unsupervised model adaptation method is performed on the universal acoustic models to recognize the speech of a specific speaker with mispronunciations and Taiwanese accent. A dynamic sentence selection algorithm, considering the mutual information of the related mispronunciations, is proposed to choose a sentence containing the most undetected mispronunciations in order to quickly extract personalized mispronunciations. The experimental results show that the proposed unsupervised adaptation approach obtains an accuracy improvement of about 2.1{\%} on the recognition of Taiwanese-accented English speech.",
author = "Chung-Hsien Wu and Su, {Hung Yu} and Liu, {Chao Hong}",
year = "2012",
month = "12",
day = "1",
doi = "10.4018/978-1-4666-1830-5.ch002",
language = "English",
isbn = "9781466618305",
pages = "12--30",
booktitle = "Multidisciplinary Computational Intelligence Techniques",
publisher = "IGI Global",

}

Wu, C-H, Su, HY & Liu, CH 2012, Efficient pronunciation assessment of taiwanese-accented english based on unsupervised model adaptation and dynamic sentence selection. 於 Multidisciplinary Computational Intelligence Techniques: Applications in Business, Engineering, and Medicine. IGI Global, 頁 12-30. https://doi.org/10.4018/978-1-4666-1830-5.ch002

Efficient pronunciation assessment of taiwanese-accented english based on unsupervised model adaptation and dynamic sentence selection. / Wu, Chung-Hsien; Su, Hung Yu; Liu, Chao Hong.

Multidisciplinary Computational Intelligence Techniques: Applications in Business, Engineering, and Medicine. IGI Global, 2012. p. 12-30.

研究成果: Chapter

TY - CHAP

T1 - Efficient pronunciation assessment of taiwanese-accented english based on unsupervised model adaptation and dynamic sentence selection

AU - Wu, Chung-Hsien

AU - Su, Hung Yu

AU - Liu, Chao Hong

PY - 2012/12/1

Y1 - 2012/12/1

N2 - This chapter presents an efficient approach to personalized pronunciation assessment of Taiwaneseaccented English. The main goal of this study is to detect frequently occurring mispronunciation patterns of Taiwanese-accented English instead of scoring English pronunciations directly. The proposed assessment help quickly discover personalized mispronunciations of a student, thus English teachers can spend more time on teaching or rectifying students' pronunciations. In this approach, an unsupervised model adaptation method is performed on the universal acoustic models to recognize the speech of a specific speaker with mispronunciations and Taiwanese accent. A dynamic sentence selection algorithm, considering the mutual information of the related mispronunciations, is proposed to choose a sentence containing the most undetected mispronunciations in order to quickly extract personalized mispronunciations. The experimental results show that the proposed unsupervised adaptation approach obtains an accuracy improvement of about 2.1% on the recognition of Taiwanese-accented English speech.

AB - This chapter presents an efficient approach to personalized pronunciation assessment of Taiwaneseaccented English. The main goal of this study is to detect frequently occurring mispronunciation patterns of Taiwanese-accented English instead of scoring English pronunciations directly. The proposed assessment help quickly discover personalized mispronunciations of a student, thus English teachers can spend more time on teaching or rectifying students' pronunciations. In this approach, an unsupervised model adaptation method is performed on the universal acoustic models to recognize the speech of a specific speaker with mispronunciations and Taiwanese accent. A dynamic sentence selection algorithm, considering the mutual information of the related mispronunciations, is proposed to choose a sentence containing the most undetected mispronunciations in order to quickly extract personalized mispronunciations. The experimental results show that the proposed unsupervised adaptation approach obtains an accuracy improvement of about 2.1% on the recognition of Taiwanese-accented English speech.

UR - http://www.scopus.com/inward/record.url?scp=84898207051&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84898207051&partnerID=8YFLogxK

U2 - 10.4018/978-1-4666-1830-5.ch002

DO - 10.4018/978-1-4666-1830-5.ch002

M3 - Chapter

AN - SCOPUS:84898207051

SN - 9781466618305

SP - 12

EP - 30

BT - Multidisciplinary Computational Intelligence Techniques

PB - IGI Global

ER -

Wu C-H, Su HY, Liu CH. Efficient pronunciation assessment of taiwanese-accented english based on unsupervised model adaptation and dynamic sentence selection. 於 Multidisciplinary Computational Intelligence Techniques: Applications in Business, Engineering, and Medicine. IGI Global. 2012. p. 12-30 https://doi.org/10.4018/978-1-4666-1830-5.ch002