DocumentCode
1860035
Title
An isolated-word speech recognition system using neural networks
Author
Runstein, Fernando ; Violaro, Fábio
Author_Institution
Centro de Pesquisas e Desenvolvimento de TELEBRAS, Sao Paulo, Brazil
Volume
1
fYear
1995
fDate
13-16 Aug 1995
Firstpage
550
Abstract
This paper describes an isolated-word, speaker independent, speech recognition system based on neural networks. The system was developed for a 32 words vocabulary including the digits, movement commands and arithmetic operation commands. The recognition rate achieved for a directly recorded speech, band-limited to 3.4 kHz, was greater than 96%
Keywords
multilayer perceptrons; self-organising feature maps; speech recognition; arithmetic operation commands; band-limited speech; digits; directly recorded speech; isolated-word speech recognition system; movement commands; neural networks; recognition rate; speaker independent recognition; Arithmetic; Cepstral analysis; Discrete Fourier transforms; Filtering; Linear predictive coding; Neural networks; Neurons; Speech recognition; Testing; Vocabulary;
fLanguage
English
Publisher
ieee
Conference_Titel
Circuits and Systems, 1995., Proceedings., Proceedings of the 38th Midwest Symposium on
Conference_Location
Rio de Janeiro
Print_ISBN
0-7803-2972-4
Type
conf
DOI
10.1109/MWSCAS.1995.504498
Filename
504498
Link To Document