Full expansion of context-dependent networks in large vocabulary speech recognition

Mehryar Mohri, Michael Riley, Don Hindle, Andrej Ljolje, Femando Pereira

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

Abstract

We combine our earlier approach to context-dependent network representation with our algorithm for determining weighted networks to build optimized networks for large-vocabulary speech recognition combining an n-gram language model, a pronunciation dictionary and context-dependency modeling. While fully-expanded networks have been used before in restrictive settings (medium vocabulary or no cross-word contexts), we demonstrate that our network determination method makes it practical to use fully-expanded networks also in large-vocabulary recognition with full cross-word context modeling. For the DARPA North American Business News task (NAB), we give network sizes and recognition speeds and accuracies using bigram and trigram grammars with vocabulary sizes ranging from 10000 to 160000 words. With our construction, the fully-expanded NAB context-dependent networks contain only about twice as many arcs as the corresponding language models. Interestingly, we also find that, with these networks, real-time word accuracy is improved by increasing the vocabulary size and n-gram order.

Original languageEnglish (US)
Title of host publicationProceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998
Pages665-668
Number of pages4
DOIs
StatePublished - 1998
Event1998 23rd IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998 - Seattle, WA, United States
Duration: May 12 1998May 15 1998

Publication series

NameICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume2
ISSN (Print)1520-6149

Other

Other1998 23rd IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998
CountryUnited States
CitySeattle, WA
Period5/12/985/15/98

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Full expansion of context-dependent networks in large vocabulary speech recognition'. Together they form a unique fingerprint.

  • Cite this

    Mohri, M., Riley, M., Hindle, D., Ljolje, A., & Pereira, F. (1998). Full expansion of context-dependent networks in large vocabulary speech recognition. In Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998 (pp. 665-668). [675352] (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2). https://doi.org/10.1109/ICASSP.1998.675352