A fast segmental Viterbi algorithm for large vocabulary recognition

Author

Laface, P. ; Vair, C. ; Fissore, L.

Author_Institution

Dipartimento di Autom. e Inf., Politecnico di Torino, Italy

Volume

1

fYear

1995

fDate

9-12 May 1995

Firstpage

560

Abstract

The paper presents a fast segmental Viterbi algorithm. A new search strategy particularly effective for very large vocabulary word recognition. It performs a tree based, time synchronous, left-to-right beam search that develops time-dependent acoustic and phonetic hypotheses. At any given time, it makes active a sub-word unit associated to an arc of a lexical tree only if that time is likely to be the boundary between the current and the next unit. This new technique, tested with a vocabulary of 188892 directory entries, achieves the same results obtained with the Viterbi algorithm, with a 35% speedup. Results are also presented for a 718 word, speaker independent continuous speech recognition task

Keywords

acoustic signal processing; maximum likelihood estimation; speech processing; speech recognition; tree searching; vocabulary; Viterbi algorithm; arc; directory entries; fast segmental Viterbi algorithm; large vocabulary word recognition; left-to-right beam search; lexical tree; search strategy; speaker independent continuous speech recognition; speedup; sub-word unit; time synchronous search; time-dependent acoustic hypothesis; time-dependent phonetic hypothesis; tree based search; Acoustic beams; Acoustic testing; Decoding; Delay effects; Laboratories; Speech recognition; Telecommunications; Telephony; Viterbi algorithm; Vocabulary;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 1995. ICASSP-95., 1995 International Conference on

Conference_Location

Detroit, MI

ISSN

1520-6149

Print_ISBN

0-7803-2431-5

Type

conf

DOI

10.1109/ICASSP.1995.479659

Filename

479659