Please use this identifier to cite or link to this item:
Title: A two-channel training algorithm for hidden markov model and its application to lip reading
Authors: Dong, L.
Foo, S.W.
Lian, Y. 
Keywords: Discriminative training
Separable-distance function
Two-channel hidden markov model
Viseme recognition
Issue Date: 1-Jun-2005
Citation: Dong, L., Foo, S.W., Lian, Y. (2005-06-01). A two-channel training algorithm for hidden markov model and its application to lip reading. Eurasip Journal on Applied Signal Processing 2005 (9) : 1382-1399. ScholarBank@NUS Repository.
Abstract: Hidden Markov model (HMM) has been a popular mathematical approach for sequence classification such as speech recognition since 1980s. In this paper, a novel two-channel training strategy is proposed for discriminative training of HMM. For the proposed training strategy, a novel separable-distance function that measures the difference between a pair of training samples is adopted as the criterion function. The symbol emission matrix of an HMM is split into two channels: a static channel to maintain the validity of the HMM and a dynamic channel that is modified to maximize the separable distance. The parameters of the two-channel HMM are estimated by iterative application of expectation-maximization (EM) operations. As an example of the application of the novel approach, a hierarchical speaker-dependent visual speech recognition system is trained using the two-channel HMMs. Results of experiments on identifying a group of confusable visemes indicate that the proposed approach is able to increase the recognition accuracy by an average of 20% compared with the conventional HMMs that are trained with the Baum-Welch estimation. © 2005 Hindawi Publishing Corporation.
Source Title: Eurasip Journal on Applied Signal Processing
ISSN: 11108657
DOI: 10.1155/ASP.2005.1382
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.


checked on Oct 15, 2018


checked on Oct 15, 2018

Page view(s)

checked on Oct 13, 2018

Google ScholarTM



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.