Dynamic Bayesian network inversion for robust speech recognition

Lei Xie, Hongwu Yang

Research output: Contribution to journalArticlepeer-review

1 Scopus citations

Abstract

This paper presents an inversion algorithm for dynamic Bayesian networks towards robust speech recognition, namely DBNI, which is a generalization of hidden Markov model inversion (HMMI). As a dual procedure of expectation maximization (EM)-based model reestimation, DBNI finds the 'uncontaminated' speech by moving the input noisy speech to the Gaussian means under the maximum likelihood (ML) sense given the DBN models trained on clean speech. This algorithm can provide both the expressive advantage from DBN and the noise-removal feature from model inversion. Experiments on the Aurora 2.0 database show that the hidden feature model (a typical DBN for speech recognition) with the DBNI algorithm achieves superior performance in terms of word error rate reduction.

Original languageEnglish
Pages (from-to)1117-1120
Number of pages4
JournalIEICE Transactions on Information and Systems
VolumeE90-D
Issue number7
DOIs
StatePublished - Jul 2007

Keywords

  • Dynamic Bayesian network
  • Hidden Markov model
  • Speech recognition

Fingerprint

Dive into the research topics of 'Dynamic Bayesian network inversion for robust speech recognition'. Together they form a unique fingerprint.

Cite this