DocumentCode :
1106578
Title :
Network-based isolated digit recognition using vector quantization
Author :
Kopec, Gary E. ; Bush, Marcia A.
Author_Institution :
Schlumberger Computer Aided Systems Research, Palo Alto, CA
Volume :
33
Issue :
4
fYear :
1985
fDate :
8/1/1985 12:00:00 AM
Firstpage :
850
Lastpage :
867
Abstract :
This paper describes a network-based approach to speaker-independent digit recognition. The digits are modeled by a pronunciation network whose arcs represent classes of acoustic-phonetic segments. Each arc is associated with a matcher for rating an input speech interval as an example of the corresponding segment class. The matchers are based on vector quantization of LPC spectra. Recognition involves finding a minimum quantization distortion path through the network by dynamic programming. The system has been evaluated in an extensive series of speaker-independent isolated digit (one-nine, oh and zero) recognition experiments using a 225-talker. multidialect database developed by Texas Instruments (TI). The best recognizer configurations achieved accuracies of 97-99 percent on the TI database.
Keywords :
Acoustic testing; Databases; Dynamic programming; Hidden Markov models; Instruments; Linear predictive coding; Pattern matching; Speech; Vector quantization; Vocabulary;
fLanguage :
English
Journal_Title :
Acoustics, Speech and Signal Processing, IEEE Transactions on
Publisher :
ieee
ISSN :
0096-3518
Type :
jour
DOI :
10.1109/TASSP.1985.1164652
Filename :
1164652
Link To Document :
بازگشت