CONFERENCE PROCEEDINGS
Advanced Search
Home > Proceedings > Volume 12707 > Article
Paper
8 June 2023 Detection of L2 Mandarin vowel pronunciation errors based on multi-task learning and articulatory features
Yizhi Wu, Ran Ji
Author Affiliations +
Proceedings Volume 12707, International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2023); 1270757 (2023) https://doi.org/10.1117/12.2680963
Event: International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2023), 2023, Changsha, China
Abstract
Computer-assisted pronunciation training (CAPT) can meet the needs of second language learners for pronunciation training. Since articulatory features can reflect the intrinsic logic of the pronunciation process and show the specific process of pronunciation, it is of great significance to incorporate them into the CAPT system. To address the complexity of vowels in continuous Mandarin speech, we established a multi-task pronunciation feature recognition model that includes four pronunciation channels. Multiple acoustic features were used to train the multitask articulatory feature recognition model on the standard Chinese corpus, obtained the correlation between multiple pronunciation features, and the trained model was then applied to the detection of pronounciation errors for L2 learners. The experimental results of Biaobei corpus show that the combination of CNN-BLSTM and multi-task training can improve the recognition accuracy of various articulatory features by 0.1% to 3.19%. The model for detecting Mandarin vowel pronunciation errors and deviations was tested on samples of second language Mandarin speakers, and it was proven to be effective in detecting errors and deviations in Mandarin vowel pronunciation.
© (2023) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Citation Download Citation
Yizhi Wu and Ran Ji "Detection of L2 Mandarin vowel pronunciation errors based on multi-task learning and articulatory features", Proc. SPIE 12707, International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2023), 1270757 (8 June 2023); https://doi.org/10.1117/12.2680963
ACCESS THE FULL ARTICLE
ORGANIZATIONAL
Sign in with credentials provided by your organization.
INSTITUTIONAL
Select your institution to access the SPIE Digital Library.
SELECT YOUR INSTITUTION
PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.
PERSONAL SIGN IN
No SPIE Account? Create one
PURCHASE THIS CONTENT
SUBSCRIBE TO DIGITAL LIBRARY
50 downloads per 1-year subscription
Members: $195
Non-members: $335 ADD TO CART
25 downloads per 1 - year subscription
Members: $145
Non-members: $250 ADD TO CART
PURCHASE SINGLE ARTICLE
Includes PDF, HTML & Video, when available
Members: $17.00
Non-members: $21.00 ADD TO CART
PROCEEDINGS
 6 PAGES
DOWNLOAD PAPER SAVE TO MY LIBRARY
GET CITATION
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Education and training
Data modeling
Tongue
Acoustics
Error analysis
Speech recognition
RELATED CONTENT
Subscribe to Digital Library
Receive Erratum Email Alert
Back to Top