Perception of terminal fall contours in speech produced by deaf persons.

Eight deaf children produced each of nine sentences. F0 measures were obtained at several locations within each utterance (starting F0, peak F0, peak F0 in the final syllable, and final F0). The relative timing of each F0 measure (ms from onset of the utterance) was also determined. In addition, several difference measures were derived. Listeners experienced with the speech of the deaf were asked to judge whether they heard a terminal fall, rise, or a flat final intonation contour in each utterance. A multiple linear regression analysis was used to determine if any combination of the acoustic measures could predict listeners' responses. The only variable that made a significant contribution to the regression function was the temporal interval between the terminal peak F0 and the final F0. That is, the more slowly the contour fell the more likely listeners were to perceive the contour as flat, regardless of the amount (in Hz or percentage F0) by which it fell. The regression equation accounted for a statistically significant but not large proportion of the total variance. This suggests that other variables, not measured in this study, play an important role in the perception of utterance final intonation contours in the speech of the deaf.