LVQ-based shift-tolerant phoneme recognition

Author

McDermott, Erik ; Katagiri, Shigeru

Author_Institution

ATR Visual Perception Res. Labs., Kyoto, Japan

Volume

39

Issue

6

fYear

1991

fDate

6/1/1991 12:00:00 AM

Firstpage

1398

Lastpage

1411

Abstract

A shift-tolerant neural network architecture for phoneme recognition is described. The system is based on algorithms for learning vector quantization (LVQ), recently developed by Kohonen (1986, 1988), which pay close attention to approximating optimal decision lines in a discrimination task. Recognition performances in the 98%-99% correct range were obtained for LVQ networks aimed at speaker-dependent recognition of phonemes in small but ambiguous Japanese phonemic classes. A correct recognition rate of 97.7% was achieved by a large LVQ network covering all Japanese consonants. These recognition results are as good as those obtained in the time delay neural network system developed by Waibel et al. (1989), and suggest that LVQ could be the basis for a high-performance speech recognition system

Keywords

data compression; encoding; learning systems; neural nets; speech recognition; Japanese consonants; Japanese phonemic classes; LVQ-based shift-tolerant phoneme recognition; discrimination task; learning vector quantization; optimal decision lines; performances; shift-tolerant neural network architecture; speech recognition; Euclidean distance; Probability density function;

fLanguage

English

Journal_Title

Signal Processing, IEEE Transactions on

Publisher

ieee

ISSN

1053-587X

Type

jour

DOI

10.1109/78.136545

Filename

136545