D. H. Whalen
Haskins Labs., 270 Crown St., New Haven, CT 06511
In a study reported at a previous meeting of the Society, subjects produced longer durations when reading lists of infrequent words than ones with frequent words. This was in spite of the fact that the words contained exactly the same phonemes by virtue of being homophones (e.g., ``right'' and ``rite''). The present report examines those differences more closely. Each major acoustic component of a word was measured. The consonantal segments included (1) the fricative noise of fricatives, (2) the resonance of nasals, liquids, and glides, and (3) a single measure for stops that included closure, burst, and aspiration. Vocalic segments were also measured. The results show that the duration for the components of infrequent words were all longer than those for frequent words, mirroring the results for overall duration. Expressed as a percentage of the total duration, though, only resonants in syllable onsets showed any difference due to frequency. Thus, in general, the duration differences resemble interspeaker, habitual rate differences more than they do effects of lexical stress or intraspeaker rate changes. These results suggest that the pattern of intraword coordination is similar for high- and low-frequency words, and that the duration differences are the result of the production process itself. [Work supported by NIH Grant DC-00825.]