Bayesian linear regression for hidden Markov model based on optimizing variational bounds

Shinji Watanabe, Atsushi Nakamura, Biing Hwang Juang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

13 Citations (Scopus)

Abstract

Linear regression for Hidden Markov Model (HMM) parameters is widely used for the adaptive training of time series pattern analysis especially for speech processing. This paper realizes a fully Bayesian treatment of linear regression for HMMs by using variational techniques. This paper analytically derives the variational lower bound of the marginalized log-likelihood of the linear regression. By using the variational lower bound as an objective function, we can optimize the model topology and hyper-parameters of the linear regression without controlling them as tuning parameters; thus, we realize linear regression for HMM parameters in a non-parametric Bayes manner. Experiments on large vocabulary continuous speech recognition confirm the generalizability of the proposed approach, especially for small quantities of adaptation data.

Original languageEnglish
Title of host publication2011 IEEE International Workshop on Machine Learning for Signal Processing - Proceedings of MLSP 2011
DOIs
Publication statusPublished - 2011 Dec 5
Externally publishedYes
Event21st IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2011 - Beijing, China
Duration: 2011 Sep 182011 Sep 21

Publication series

NameIEEE International Workshop on Machine Learning for Signal Processing

Other

Other21st IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2011
CountryChina
CityBeijing
Period11/9/1811/9/21

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Signal Processing

Fingerprint Dive into the research topics of 'Bayesian linear regression for hidden Markov model based on optimizing variational bounds'. Together they form a unique fingerprint.

  • Cite this

    Watanabe, S., Nakamura, A., & Juang, B. H. (2011). Bayesian linear regression for hidden Markov model based on optimizing variational bounds. In 2011 IEEE International Workshop on Machine Learning for Signal Processing - Proceedings of MLSP 2011 [6064605] (IEEE International Workshop on Machine Learning for Signal Processing). https://doi.org/10.1109/MLSP.2011.6064605