Evaluation and integration of acoustic features in speech perception

Publisher Website
Google Scholar
Abstract
Identification of synthetic stop consonants as /bae/, /pae/, /dae/ or /tae/ was examined where the stimuli varied independently on voice onset time (VOT), the consonantal 2nd and 3rd formant (F2-F3) transitions, and the intensity of the aspiration noise during the VOT period. The patterns of the identification probabilities were complex, but systematic, functions of each of the independent variables. The likelihood of identifying a stimulus as /bae/ or /pae/, rather than /dae/ or /tae/, was strongly influenced by the VOT and by the F2-F3 transitions. The likelihood of identifying a stimulus as /bae/ or /dae/, rather than /pae/ or /tae/, depended on the F2-F3 transitions and VOT. Three explanations of these results were considered within a fuzzy logical model of human speech perception: that there is an interaction in the evaluation of acoustic features: that the listener requires more extreme values of acoustic features for some speech sounds than for that of other speech sounds; and that the aspiration noise during the VOT period serves as an independent acoustic feature to distinguish /pae/ and /bae/ from /tae/ and /dae/.

This publication has 1 reference indexed in Scilit: