Imitation of synthesized vowels by children and adults

Abstract

Listen

Translate article

Imitations of 15 synthesized vowels, some English vowels and some not, were obtained from ten adults (five men and five women) and ten six-year-old children (five boys and five girls). Estimates of the first three formant frequencies, F1, F2, and F3, were made from spectrograms of the vowel imitations. The reliability of reproduction was assessed by calculating standard deviations for five imitations each of ten of the synthesized vowels. Generally, both the intra- and intersubject variabilities of the formant frequencies were great for children than adults, but the differences in intra-subject variability were not much greater than the differences in measurement error associated with different fundamental frequencies. Subjects tended to reproduce non-English vowels less reliably than English vowels, although adults were influenced less by phonetic familiarity than were the children. Vowel familiarity appeared to be especially important for reliable reproduction of the F2 frequency. Plotting of the data for English vowels in a F1−F2 plane with linear dimensions revealed an almost linear clustering for the four subject groups, but the group clustering was not as linear for the non-English vowels. [Work supported by NIH.]