Vowel recognition in Romanian language

Author

Grigore, Ovidiu ; Gavat, Inge ; Zirra, Matei

Author_Institution

Fac. of Electron. & Telecommun., Univ. Politehnica Bucharest, romania

Volume

2

fYear

1997

fDate

7-11 Oct 1997

Firstpage

589

Abstract

In this paper are presented results obtained in a vowel recognition task applying unsupervised and supervised fuzzy algorithms and neural networks. The vowels, uttered from 10 speakers each in 250 different contexts are recognized using as features the first three formant frequencies. After an introduction, the feature extraction is presented and then the two fuzzy algorithms (fuzzy ISODATA, fuzzy k-NN) and the two neural nets (nonlinear perceptron, Kohonen map) used for recognition are given

Keywords

feature extraction; fuzzy neural nets; multilayer perceptrons; pattern classification; self-organising feature maps; speech recognition; unsupervised learning; Kohonen map; Romanian language; feature extraction; first three formant frequencies; fuzzy ISODATA algorithm; fuzzy k-NN algorithm; homomorphic analysis; neural networks; nonlinear perceptron; pattern classification; speech recognition; supervised fuzzy algorithms; unsupervised fuzzy algorithms; vowel recognition; Automatic speech recognition; Cepstrum; Clustering algorithms; Feature extraction; Frequency; Fuzzy neural networks; Histograms; Neural networks; Prototypes; Speech recognition;

fLanguage

English

Publisher

ieee

Conference_Titel

Semiconductor Conference, 1997. CAS '97 Proceedings., 1997 International

Conference_Location

Sinaia

Print_ISBN

0-7803-3804-9

Type

conf

DOI

10.1109/SMICND.1997.651330

Filename

651330