APSIPA Transactions on Signal and Information Processing > Vol 1 > Issue 1

Bayesian approaches to acoustic modeling: a review

Shinji Watanabe, Mitsubishi Electric Research Laboratories (MERL), USA, watanabe@merl.com , Atsushi Nakamura, NTT Communication Science Laboratories, Japan
 
Suggested Citation
Shinji Watanabe and Atsushi Nakamura (2012), "Bayesian approaches to acoustic modeling: a review", APSIPA Transactions on Signal and Information Processing: Vol. 1: No. 1, e5. http://dx.doi.org/10.1017/ATSIP.2012.6

Publication Date: 06 Dec 2012
© 2012 Shinji Watanabe and Atsushi Nakamura
 
Subjects
 
Keywords
Speech processingMachine learningBayesian approachApproximate Bayesian inference
 

Share

Open Access

This is published under the terms of the Creative Commons Attribution licence.

Downloaded: 617 times

In this article:
I. INTRODUCTION 
II. MAP 
III. BIC 
IV. VB 
V. MCMC 
VI. SUMMARY AND FUTUREPERSPECTIVE 

Abstract

This paper focuses on applications of Bayesian approaches to acoustic modeling for speech recognition and related speech-processing applications. Bayesian approaches have been widely studied in the fields of statistics and machine learning, and one of their advantages is that their generalization capability is better than that of conventional approaches (e.g., maximum likelihood). On the other hand, since inference in Bayesian approaches involves integrals and expectations that are mathematically intractable in most cases and require heavy numerical computations, it is generally difficult to apply them to practical speech recognition problems. However, there have been many such attempts, and this paper aims to summarize these attempts to encourage further progress on Bayesian approaches in the speech-processing field. This paper describes various applications of Bayesian approaches to speech processing in terms of the four typical ways of approximating Bayesian inferences, i.e., maximum a posteriori approximation, model complexity control using a Bayesian information criterion based on asymptotic approximation, variational approximation, and Markov chain Monte Carlo-based sampling techniques.

DOI:10.1017/ATSIP.2012.6