Discriminative GMM-HMM acoustic model selection using two-level bayesian ying-yang harmony learning
Refereed conference paper presented and published in conference proceedings


Times Cited
Altmetrics Information
.

Other information
AbstractThis paper proposes a two-level Bayesian Ying-Yang (BYY) harmony learning based acoustic model discriminative training method. In this method, a rival penalized competitive learning (RPCL) simplified BYY harmony learning based discriminative training is conducted at the HMM state level to optimizing the state boundaries, while a BYY based model selection is conducted at the Gaussian mixture components level to determine the Gaussian mixture components within the same HMM state. Two levels of learning work coordinately and have good convergence. Experiments show that the trained model is more discriminative with better recognition performance, and also more compact with smaller number of Gaussian components. © Springer-Verlag 2013.
All Author(s) ListPang Z., Tu S., Wu X., Xu L.
Name of Conference3rd Sino-Foreign-Interchange Workshop on Intelligent Science and Intelligent Data Engineering, IScIDE 2012
Start Date of Conference15/10/2012
End Date of Conference17/10/2012
Place of ConferenceNanjing
Country/Region of ConferenceChina
Detailed descriptioned. by J. Yang, F. Fang, and C. Sun.
Year2013
Month12
Day1
Volume Number7751 LNCS
PublisherSpringer Verlag
Place of PublicationGermany
Pages719 - 726
ISBN9783642366680
ISSN0302-9743
LanguagesEnglish-United Kingdom
KeywordsBayesian Ying-Yang harmony learning, Discriminative training, Hidden Markov model, Large vocabulary continuous speech recognition, Rival penalized competitive learning

Last updated on 2020-14-10 at 02:22