An acoustic correlate of syllabicity in english

Abstract

The purpose of this research was to test three hypotheses (1) Increasing the duration of a nonsyllabic segment will cause the segment to be perceived as syllabic. (2) The perception of syllabicity is categorical. (3) Listeners will differ in their labeling responses to stimuli when told to judge them in different speech styles. Seven words were synthesized: “blow,” “plight,” “dress,” “crest,” “prayed,” “broke,” and “sport.” The steady-state portion of the /l/ or the /r/ (for “sport,” aspiration Of /p/) was lengthened in 10-msec increments. Subjects were asked to decide whether the word was monosyllabic or disyllabic: “blow-below,” “plight-polite,” “dress-duress,” “crest-caressed,” “prayed-parade,” “broke-baroque,” and “sport-support.” Five groups of 15 listeners each participated in the labeling task. Four groups heard the stimuli with a precursor frame, “The word you will hear next is ___” in one of four speech styles: formal-slow, formal-fast, casual-slow, casual-fast. Each group was told to judge the words according to the criterion of being spoken in one of the designated styles. The fifth group heard the stimuli with no precursor and received no style instructions. Results showed that (1) increased durations of /r/ and /l/ did result in perception of the words as disyllabic, but increasing the duration of aspiration in “sport” did not; (2) perception of syllabicity appears to be categorical; and (3) there were no statistical differences among the four speech styles. [Work supported by NSF SOC75-10043.]