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.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Dean LUO, Yu QIAO, Nobuaki MINEMATSU, Keikichi HIROSE, "Regularized Maximum Likelihood Linear Regression Adaptation for Computer-Assisted Language Learning Systems" in IEICE TRANSACTIONS on Information,
vol. E94-D, no. 2, pp. 308-316, February 2011, doi: 10.1587/transinf.E94.D.308.
Abstract: 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.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E94.D.308/_p
Copy
@ARTICLE{e94-d_2_308,
author={Dean LUO, Yu QIAO, Nobuaki MINEMATSU, Keikichi HIROSE, },
journal={IEICE TRANSACTIONS on Information},
title={Regularized Maximum Likelihood Linear Regression Adaptation for Computer-Assisted Language Learning Systems},
year={2011},
volume={E94-D},
number={2},
pages={308-316},
abstract={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.},
keywords={},
doi={10.1587/transinf.E94.D.308},
ISSN={1745-1361},
month={February},}
Copy
TY - JOUR
TI - Regularized Maximum Likelihood Linear Regression Adaptation for Computer-Assisted Language Learning Systems
T2 - IEICE TRANSACTIONS on Information
SP - 308
EP - 316
AU - Dean LUO
AU - Yu QIAO
AU - Nobuaki MINEMATSU
AU - Keikichi HIROSE
PY - 2011
DO - 10.1587/transinf.E94.D.308
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E94-D
IS - 2
JA - IEICE TRANSACTIONS on Information
Y1 - February 2011
AB - 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.
ER -