HKUST Library Institutional Repository Banner

HKUST Institutional Repository >
Computer Science and Engineering >
CSE Journal/Magazine Articles >

Please use this identifier to cite or link to this item: http://hdl.handle.net/1783.1/6546
Title: Maximum penalized likelihood kernel regression for fast adaptation
Authors: Mak, Brian Kan-Wing
Lai, Tsz-Chung
Tsang, I. W.
Kwok, James Tin-Yau
Keywords: Kernel regression
Maximum-likelihood linear regression (MLLR)
Reference speaker weighting
Speaker adaptation
Issue Date: Sep-2009
Citation: IEEE transactions on audio, speech, and language processing, v. 17, iss. 7, September 2009, p. 1372-1381
Abstract: This paper proposes a nonlinear generalization of the popular maximum-likelihood linear regression (MLLR) adaptation algorithm using kernel methods. The proposed method, called maximum penalized likelihood kernel regression adaptation (MPLKR), applies kernel regression with appropriate regularization to determine the affine model transform in a kernel-induced high-dimensional feature space. Although this is not the first attempt of applying kernel methods to conventional linear adaptation algorithms, unlike most of other kernelized adaptation methods such as kernel eigenvoice or kernel eigen-MLLR, MPLKR has the advantage that it is a convex optimization and its solution is always guaranteed to be globally optimal. In fact, the adapted Gaussian means can be obtained analytically by simply solving a system of linear equations. From the Bayesian perspective, MPLKR can also be considered as the kernel version of maximum a posteriori linear regression (MAPLR) adaptation. Supervised and unsupervised speaker adaptation using MPLKR were evaluated on the Resource Management and Wall Street Journal 5K tasks, respectively, achieving a word error rate reduction of 23.6% and 15.5% respectively over the speaker-independently model.
Rights: © 2009 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE. This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.
URI: http://hdl.handle.net/1783.1/6546
Appears in Collections:CSE Journal/Magazine Articles

Files in This Item:

File Description SizeFormat
MaximumPenalized.pdf562KbAdobe PDFView/Open

Find published version via OpenURL Link Resolver

All items in this Repository are protected by copyright, with all rights reserved.