Noisy speech recognition by using variance adapted hidden Markov models

Jen-Tzung Chien, Lee Min Lee, Hsiao Chuan Wang

Research output: Contribution to journalArticle

4 Scopus citations

Abstract

Shrinkage of variances in HMM due to additive noise is an important factor in speech recognition process. This variance shrinkage can be nonparametrically estimated from a training database. In the Letter we take this factor into account in the projection measure for noisy speech recognition. Experiments show that recognition rates can be improved significantly.

Original languageEnglish
Pages (from-to)1555-1556
Number of pages2
JournalElectronics Letters
Volume31
Issue number18
DOIs
StatePublished - 31 Aug 1995

Keywords

  • Hidden Markov models
  • Speech recognition

Fingerprint Dive into the research topics of 'Noisy speech recognition by using variance adapted hidden Markov models'. Together they form a unique fingerprint.

  • Cite this