Pseudo-segment based speech recognition using neural recurrent whole-word recognizers

Author

Le Cerf, Philippe ; Demuynck, Kris ; Duchateau, Jacques ; Van Compernolle, Dirk

Author_Institution

ESAT, Katholieke Univ., Leuven, Heverlee, Belgium

Volume

i

fYear

1994

fDate

19-22 Apr 1994

Abstract

Describes a recurrent neural network based, isolated word speech recognizer. The recognizer uses 2 MLPs. A first, static MLP is used for classification of frames in phonemes. Next, a time compression step is applied. The resulting pseudo-segments are then used as inputs for a second, dynamic MLP that integrates the information over time to decide the current word. The authors apply this approach on an isolated digit recognition task and compare the results with hybrid MLP/HMM approach using the same static MLP

Keywords

multilayer perceptrons; recurrent neural nets; speech recognition; frames; isolated digit recognition task; isolated word speech recognizer; multilayer perceptron; neural recurrent whole-word recognizers; phoneme; pseudosegment based speech recognition; recurrent neural network; time compression step; Hidden Markov models; Multilayer perceptrons; Neural networks; Pattern classification; Prototypes; Recurrent neural networks; Speech recognition; Testing; Thumb; Vocabulary;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 1994. ICASSP-94., 1994 IEEE International Conference on

Conference_Location

Adelaide, SA

ISSN

1520-6149

Print_ISBN

0-7803-1775-0

Type

conf

DOI

10.1109/ICASSP.1994.389220

Filename

389220