Year: 2014
Type: article
Abstract: This paper presents a mathematical model of formant transitions in CV syllables of the form f(t)=fv +[(fc -fv)+ct/τ]exp(-t/τ) where f(t) = frequency at time t fv = frequency of vowel target fc = frequency of formant onset c = the amount of perturbation from an exponential decay, such that the initial slope is zeroforc=fc -fv. τ = a time constant. This model is based on the exponential decay model of Broad & Clermont (1987), but incorporates the additional factor c, which describes perturbation from a simple exponential decay. Such perturbation may arise from secondary articulation such as palatalization. The model was tested on utterances of isolated syllables consisting of the Russian voiced stops /b d g/ and their palatalized counterparts /b’ d’ g’/, each followed by the vowels /i e a o u/. Formant frequencies were measured at each glottal pulse for the first 300 milliseconds from release, together with the time from release. Curves were then fitted to the measured data and values computed for each of the parameters in the model. Analyses of variance performed on the computed parameter values showed that to a large extent the parameter values of the model reflect the phonetic categories vowel, consonant place of (primary) articulation and palatalization.
Author: Stephen M. Howie
Language: English
Subfield: Experimental and Cognitive Psychology
Field: Psychology
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W1162719865