Static, dynamic, and relational factors in vowel perception

Abstract

This paper reviews theories and empirical findings that bear on the perception of English vowels, with emphasis on the comparison of data-analytic pattern recognition studies with results from speech perception experiments. Methodological approaches and empirical results from research at the University of Alberta [e.g., T. M. Nearey and P. F. Assmann, J. Acoust. Soc. Am. 80, 1297–1308 (1986)] will be compared and contrasted to work in other laboratories. Issues to be considered include: (1) the role F1 and F2 and that of fundamental frequency and higher formants; (2) “vowel-space and speaking rate normalization” (global speaker context); (3) vowel-inherent spectral change (diphthongizatiun); and (4) local phonetic context (e.g., consonantal context). While all of the above factors have been shown to have reliable effects on vowel perception, the relative weight of such effects and the circumstances which may alter these weights remain far from clear. Analysis of these problems in current research and suggestions for future studies will be presented.