Regularized Maximum Likelihood Linear Regression Adaptation for Computer-Assisted Language Learning Systems

Dean LUO  Yu QIAO  Nobuaki MINEMATSU  Keikichi HIROSE  

Publication
IEICE TRANSACTIONS on Information and Systems   Vol.E94-D   No.2   pp.308-316
Publication Date: 2011/02/01
Online ISSN: 1745-1361
DOI: 10.1587/transinf.E94.D.308
Print ISSN: 0916-8532
Type of Manuscript: PAPER
Category: Educational Technology
Keyword: 
CALL,  speaker adaptation,  MLLR,  goodness of pronunciation,  automatic scoring,  pronunciation evaluation,  error detection,  

Full Text: PDF(571KB)>>
Buy this Article




Summary: 
This study focuses on speaker adaptation techniques for Computer-Assisted Language Learning (CALL). We first investigate the effects and problems of Maximum Likelihood Linear Regression (MLLR) speaker adaptation when used in pronunciation evaluation. Automatic scoring and error detection experiments are conducted on two publicly available databases of Japanese learners' English pronunciation. As we expected, over-adaptation causes misjudgment of pronunciation accuracy. Following the analysis, we propose a novel method, Regularized Maximum Likelihood Regression (Regularized-MLLR) adaptation, to solve the problem of the adverse effects of MLLR adaptation. This method uses a group of teachers' data to regularize learners' transformation matrices so that erroneous pronunciations will not be erroneously transformed as correct ones. We implement this idea in two ways: one is using the average of the teachers' transformation matrices as a constraint to MLLR, and the other is using linear combinations of the teachers' matrices to represent learners' transformations. Experimental results show that the proposed methods can better utilize MLLR adaptation and avoid over-adaptation.