模式识别与人工智能
Friday, Apr. 4, 2025 Home      About Journal      Editorial Board      Instructions      Ethics Statement      Contact Us                   中文
  2010, Vol. 23 Issue (4): 450-455    DOI:
Orignal Article Current Issue| Next Issue| Archive| Adv Search |
A Constrained Line Search Optimization Method for Discriminative Training of HMMs
LIU Cong,HU Yu,DAI Li-Rong,WANG Ren-Hua
iFlytek Speech Laboratory,Department of Electronic Engineering and Information Science,University of Science and Technology of China,Hefei 230027

Download: PDF (442 KB)   HTML (1 KB) 
Export: BibTeX | EndNote (RIS)      
Abstract  In this paper a optimization algorithm called constrained line search (CLS) is proposed for discriminative training (DT) of Gaussian mixture Continuous Density Hidden Markov Model (CDHMM) in speech recognition. The CLS method is used to optimize the objective function of Maximum Mutual Information (MMI) criterion based discriminative training. In this method, discriminative training of HMM is firstly treated as a constrained optimization problem, and Kullback-Leibler divergence (KLD) between models is explicitly applied as a constraint during optimization. Based upon the idea of line search, it is pointed out that a simple formula of HMM parameters can be expressed as a simple quadratic formula by constraining the KLD between HMM of two successive iterations. The proposed CLS method can be applied to optimize all model parameters in Gaussian mixture CDHMMs, including means, covariances, and mixture weights. The proposed CLS approach is investigated on two benchmark speech recognition tasks of both English and Chinese, including TIDIGITS and Mandarin 863 database. Experimental results show that the CLS optimization method consistently outperforms the conventional extended Baum-Welch (EBW) method in both recognition performance and convergence characteristic.
Key wordsAutomatic Speech Recognition      Discriminative Training      Constrained Line Search (CLS)     
Received: 25 June 2009     
ZTFLH: TN912.34  
Service
E-mail this article
Add to my bookshelf
Add to citation manager
E-mail Alert
RSS
Articles by authors
LIU Cong
HU Yu
DAI Li-Rong
WANG Ren-Hua
Cite this article:   
LIU Cong,HU Yu,DAI Li-Rong等. A Constrained Line Search Optimization Method for Discriminative Training of HMMs[J]. , 2010, 23(4): 450-455.
URL:  
http://manu46.magtech.com.cn/Jweb_prai/EN/      OR     http://manu46.magtech.com.cn/Jweb_prai/EN/Y2010/V23/I4/450
Copyright © 2010 Editorial Office of Pattern Recognition and Artificial Intelligence
Address: No.350 Shushanhu Road, Hefei, Anhui Province, P.R. China Tel: 0551-65591176 Fax:0551-65591176 Email: bjb@iim.ac.cn
Supported by Beijing Magtech  Email:support@magtech.com.cn