A double blind peer reviewed online publication with in-print supplement since 2010    ISSN: 1949-260X

JTCLT Abstract

Volume 9 Number 2, 2018
Full issue PDF

Wei, W., Zhang, J. (2018). An Intelligent Chinese Pronunciation Teaching App and the Preliminary Result of a Teaching Experiment. Journal of Technology and Chinese Language Teaching, 9(2), 83-97.

Full paper

Abstract/摘要:

Pronunciation has always been a challenge in the learning and teaching of Chinese as a second language. Two common best practices to improve learners’ pronunciation include providing learners with sufficient meaningful practice opportunities and quick feedback from their instructors. However, due to limited classroom time and one-to-many relationship between teachers and students, it is difficult to follow the above two practices in classroom instruction. Driven by artificial intelligence technology, intelligent computer assisted pronunciation teaching (CAPT) system can help learners to practice pronunciation on their own and receive instant feedback. This paper reports the findings of a six-week teaching experiment in using Erya Chinese pronunciation APP at Beijing Language and Culture University. Our preliminary results showed that the total number of pronunciation exercises was 28,101 times, with an average of 780.6 times per student, of which the number of after-school exercises accounted for 91.9% in quantity; under the guidance of detailed feedback information, students can remedy 83.2% of their pronunciation errors efficiently by only 1.5 exercises on average. This result showed that the intelligent CAPT system can help the learner to complete a large number of pronunciation exercises and improve the pronunciation quality.

发音教学一直是汉语作为第二语言教学中的难点问题。近年来迅速发展的人工智能发音教学技术(CAPT)能够自动诊断学生发音中的偏误,并给出正音的引导信息,从而帮助学生不受时空限制的完成自主发音学习,被认为是克服二语发音教学问题的希望技术。但是由于出现时间尚短,技术也有待完善,其教学辅助效果到底如何是一个令语学界感兴趣的问题。北京语言大学智能语音习得技术实验室一直以来从事汉语作为第二语言的智能语音教学技术的研究,开发出以尔雅中文教材为模板的汉语CAPT-APP软件。为了探索CAPT实效问题,本文以该软件为工具,在北京语言大学的两个留学生班级进行了为期6周的教学实验。初步的实验结果显示,36名学生在6周的时间内使用APP练习发音的总次数为28101次,平均每名学生练习次数为780.6次,其中,课后练习次数占到了91.9%;发音偏误细节反馈信息能够帮助学生改善83.2%的发音问题,并且改善一个发音问题平均只需要约1.5次的重复练习。这样的结果说明CAPT教学工具对于二语语音教学的确非常有效,不仅能够有效辅助学习者完成大量发音练习,而且有助于他们改善发音质量。

This website is supported by
The TCLT Standing Committee
Department of World Languages, Literatrues, and Cultures, Middle Tennessee State University
Department of Chinese Language and Literature, University of Macau
Page last updated: 2016-12-31