TY - JOUR
T1 - A discriminative training algorithm for hidden Markov models
AU - Ben-Yishai, Assaf
AU - Burshtein, David
N1 - Funding Information:
Manuscript received August 9, 2001; revised October 7, 2003. This research was supported by the KITE consortium of the Israeli Ministry of Industry and Trade. The associate editor coordinating the review of this manuscript and approving it for publication was Dr. Andreas Stolcke.
PY - 2004/5
Y1 - 2004/5
N2 - We introduce a discriminative training algorithm for the estimation of hidden Markov model (HMM) parameters. This algorithm is based on an approximation of the maximum mutual information (MMI) objective function and its maximization in a technique similar to the expectation-maximization (EM) algorithm. The algorithm is implemented by a simple modification of the standard Baum-Welch algorithm, and can be applied to speech recognition as well as to word-spotting systems. Three tasks were tested: Isolated digit recognition in a noisy environment, connected digit recognition in a noisy environment and word-spotting. In all tasks a significant improvement over maximum likelihood (ML) estimation was observed. We also compared the new algorithm to the commonly used extended Baum-Welch MMI algorithm. In our tests the algorithm showed advantages in terms of both performance and computational complexity.
AB - We introduce a discriminative training algorithm for the estimation of hidden Markov model (HMM) parameters. This algorithm is based on an approximation of the maximum mutual information (MMI) objective function and its maximization in a technique similar to the expectation-maximization (EM) algorithm. The algorithm is implemented by a simple modification of the standard Baum-Welch algorithm, and can be applied to speech recognition as well as to word-spotting systems. Three tasks were tested: Isolated digit recognition in a noisy environment, connected digit recognition in a noisy environment and word-spotting. In all tasks a significant improvement over maximum likelihood (ML) estimation was observed. We also compared the new algorithm to the commonly used extended Baum-Welch MMI algorithm. In our tests the algorithm showed advantages in terms of both performance and computational complexity.
KW - Discriminative training
KW - Hidden Markov model (HMM)
KW - Maximum mutual information (MMI) criterion
UR - http://www.scopus.com/inward/record.url?scp=2442503633&partnerID=8YFLogxK
U2 - 10.1109/TSA.2003.822639
DO - 10.1109/TSA.2003.822639
M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???
AN - SCOPUS:2442503633
SN - 1063-6676
VL - 12
SP - 204
EP - 217
JO - IEEE Transactions on Speech and Audio Processing
JF - IEEE Transactions on Speech and Audio Processing
IS - 3
ER -