Recognition of vowels in continuous speech by using formants

Abstract

Speech consists of acoustic pressure waves created by the voluntary movements of anatomical structures in the human speech production system. These waveforms are broadly classified into voiced and unvoiced speech. Voiced sounds (vowels for example), produce quasi-periodic pulses of air which are acoustically filtered as they propagate through the vocal tract. The main distinction between vowels and consonants is that vowels resonate in the throat. Formants are exactly the resonant frequencies of a vocal tract when pronouncing a vowel. In this paper we attempt to carry out Vowel Recognition through Formant Analysis in Serbian language, wherein we detect which of the five Serbian vowels is spoken by the Speaker. Here we describe a standard approach for classification of vowels in continuous speech based on three formants: F1, F2 and F3. We have investigated the correlations between formants in each vowel and developed the algorithm to reduce the overlap of different vowels in F1-F2 and F2-F3 planes.