Improved lexicon formation through removal of co-articulation and acoustic recognition errors

Darryl Stewart, Ji Ming, F. J. Smith

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

Abstract

It is becoming increasingly more necessary that speech recognition systems contain an accurate lexicon, consisting of likely word pronunciations that actually occur within a given domain. Given the increasing size of speech databases, it would appear that data driven approaches are best suited to derive such pronunciations. Presently, however, such an approach often introduces implausible pronunciations, resulting in a higher degree of confusability within the decoder. In this paper, we outline a novel data driven approach which aims to improve the quality of extracted word pronunciations through the removal of co-articulation effects and acoustic model misclassifications from the speech data. A number of selection constraints are additionally employed to exclude any improbable pronunciation alternatives. Initial experiments have shown that the approach does indeed provide plausible pronunciation alternatives without introducing improbable pronunciations.

Original languageEnglish
Title of host publication6th International Conference on Spoken Language Processing, ICSLP 2000
PublisherInternational Speech Communication Association
ISBN (Electronic)7801501144, 9787801501141
Publication statusPublished - 01 Jan 2000
Event6th International Conference on Spoken Language Processing, ICSLP 2000 - Beijing, China
Duration: 16 Oct 200020 Oct 2000

Publication series

Name6th International Conference on Spoken Language Processing, ICSLP 2000

Conference

Conference6th International Conference on Spoken Language Processing, ICSLP 2000
Country/TerritoryChina
CityBeijing
Period16/10/200020/10/2000

ASJC Scopus subject areas

  • Linguistics and Language
  • Language and Linguistics

Fingerprint

Dive into the research topics of 'Improved lexicon formation through removal of co-articulation and acoustic recognition errors'. Together they form a unique fingerprint.

Cite this