Improvements of search error risk minimization in viterbi beam search for speech recognition

Takaaki Hori, Shinji Watanabe, Atsushi Nakamura

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

Abstract

This paper describes improvements in a search error risk minimization approach to fast beam search for speech recognition. In our previous work, we proposed this approach to reduce search errors by optimizing the pruning criterion. While conventional methods use heuristic criteria to prune hypotheses, our proposed method employs a pruning function that makes a more precise decision using rich features extracted from each hypothesis. The parameters of the function can be estimated to minimize a loss function based on the search error risk. In this paper, we improve this method by introducing a modified loss function, arc-averaged risk, which potentially has a higher correlation with actual error rate than the original one. We also investigate various combinations of features. Experimental results show that further search error reduction over the original method is obtained in a 100K-word vocabulary lecture speech transcription task.

Original languageEnglish
Title of host publicationProceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010
PublisherInternational Speech Communication Association
Pages1962-1965
Number of pages4
Publication statusPublished - 2010
Externally publishedYes

Publication series

NameProceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010

Keywords

  • Beam search
  • Pruning
  • Search error
  • Speech recognition
  • WFST

ASJC Scopus subject areas

  • Language and Linguistics
  • Speech and Hearing

Fingerprint Dive into the research topics of 'Improvements of search error risk minimization in viterbi beam search for speech recognition'. Together they form a unique fingerprint.

Cite this