Abstract

Twenty normal-speaking young adult males phonated (at one intensity) in modal register, with simulated abnormal vocal roughness, and in vocal fry register each of the five test vowels, /u/, /i/, /^/, /a/ and /æ/. Each test production was rated for roughness by a panel of 11 listeners; a median of the ratings for each production was then obtained. Additionally, the acoustic wave of each production was analyzed to produce a narrow-band (3 Hz) intensity-by-frequency acoustic spectrum in which multiple (25) measures of spectral noise (inharmonic component) level were obtained. For each test vowel, the median roughness ratings for fry productions were similar in magnitude to those for simulated abnormally rough productions and exceeded ( P < 0.05) those for modal productions. Further, fry and simulated abnormally rough productions did not differ with respect to spectral noise levels, but the spectral noise levels for both fry and abnormally rough productions were greater ( P < 0.05) than those for modal productions. A high degree of linear relationship was demonstrated between the roughness ratings and the spectral noise levels measures for the productions of each test vowel.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call