LARGE VOCABULARY SPEECH RECOGNITION USING A HIDDEN MARKOV MODEL FOR ACOUSTIC/PHONETIC CLASSIFICATION.

S. E. Levinson; A. Ljolje; L. G. Miller

LARGE VOCABULARY SPEECH RECOGNITION USING A HIDDEN MARKOV MODEL FOR ACOUSTIC/PHONETIC CLASSIFICATION.

S. E. Levinson, A. Ljolje, L. G. Miller

Research output: Contribution to journal › Conference article › peer-review

Abstract

Experiments with a speech recognition system are reported. The system comprises an acoustic/phonetic decoder, a lexical access mechanism and a syntax analyzer. The acoustic, phonetic and lexical processing are based on a continuously-variable-duration hidden Markov model (CVDHMM). The syntactic component is based on the Cocke-Kasami-Young (CKY) parser and a content-free covering grammar of English. Lexical items are represented in terms of the 43 phonetic units. In recognition tests conducted on a separate data set, a 70% correct recognition rate on phonetic units in fluent speech was observed. In two additional tests on isolated words, a 40% correct word recognition was observed with the complete 52,000 word lexicon. When the vocabulary size was reduced to 1040 words, the recognition rate improved to 80%. After syntax analysis the word recognition rate rose to 90%.

Original language	English (US)
Pages (from-to)	505-508
Number of pages	4
Journal	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
State	Published - 1988
Externally published	Yes

ASJC Scopus subject areas

Software
Signal Processing
Electrical and Electronic Engineering

Library availability

Discover UIUC Full Text

Cite this

@article{b0820aebc7b64bdab6cb288f602cb1b2,

title = "LARGE VOCABULARY SPEECH RECOGNITION USING A HIDDEN MARKOV MODEL FOR ACOUSTIC/PHONETIC CLASSIFICATION.",

abstract = "Experiments with a speech recognition system are reported. The system comprises an acoustic/phonetic decoder, a lexical access mechanism and a syntax analyzer. The acoustic, phonetic and lexical processing are based on a continuously-variable-duration hidden Markov model (CVDHMM). The syntactic component is based on the Cocke-Kasami-Young (CKY) parser and a content-free covering grammar of English. Lexical items are represented in terms of the 43 phonetic units. In recognition tests conducted on a separate data set, a 70% correct recognition rate on phonetic units in fluent speech was observed. In two additional tests on isolated words, a 40% correct word recognition was observed with the complete 52,000 word lexicon. When the vocabulary size was reduced to 1040 words, the recognition rate improved to 80%. After syntax analysis the word recognition rate rose to 90%.",

author = "Levinson, {S. E.} and A. Ljolje and Miller, {L. G.}",

year = "1988",

language = "English (US)",

pages = "505--508",

journal = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

issn = "0736-7791",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - LARGE VOCABULARY SPEECH RECOGNITION USING A HIDDEN MARKOV MODEL FOR ACOUSTIC/PHONETIC CLASSIFICATION.

AU - Levinson, S. E.

AU - Ljolje, A.

AU - Miller, L. G.

PY - 1988

Y1 - 1988

N2 - Experiments with a speech recognition system are reported. The system comprises an acoustic/phonetic decoder, a lexical access mechanism and a syntax analyzer. The acoustic, phonetic and lexical processing are based on a continuously-variable-duration hidden Markov model (CVDHMM). The syntactic component is based on the Cocke-Kasami-Young (CKY) parser and a content-free covering grammar of English. Lexical items are represented in terms of the 43 phonetic units. In recognition tests conducted on a separate data set, a 70% correct recognition rate on phonetic units in fluent speech was observed. In two additional tests on isolated words, a 40% correct word recognition was observed with the complete 52,000 word lexicon. When the vocabulary size was reduced to 1040 words, the recognition rate improved to 80%. After syntax analysis the word recognition rate rose to 90%.

AB - Experiments with a speech recognition system are reported. The system comprises an acoustic/phonetic decoder, a lexical access mechanism and a syntax analyzer. The acoustic, phonetic and lexical processing are based on a continuously-variable-duration hidden Markov model (CVDHMM). The syntactic component is based on the Cocke-Kasami-Young (CKY) parser and a content-free covering grammar of English. Lexical items are represented in terms of the 43 phonetic units. In recognition tests conducted on a separate data set, a 70% correct recognition rate on phonetic units in fluent speech was observed. In two additional tests on isolated words, a 40% correct word recognition was observed with the complete 52,000 word lexicon. When the vocabulary size was reduced to 1040 words, the recognition rate improved to 80%. After syntax analysis the word recognition rate rose to 90%.

UR - http://www.scopus.com/inward/record.url?scp=0023804696&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0023804696&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:0023804696

SN - 0736-7791

SP - 505

EP - 508

JO - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

JF - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

ER -

LARGE VOCABULARY SPEECH RECOGNITION USING A HIDDEN MARKOV MODEL FOR ACOUSTIC/PHONETIC CLASSIFICATION.

Abstract

ASJC Scopus subject areas

Library availability

Related links

Fingerprint

Cite this