Acoustic and Psychoacoustic Analyses of Predominant Strain Elicited via Modulated Vocal Effort.

PURPOSE: The purpose of this study was to examine objective correlates of strain by reducing the covarying effects of other voice qualities. Predominant strain was elicited in a group of typical voice individuals instructed to modulate their vocal effort. The resulting sustained vowel productions were evaluated by (a) auditory-perceptual ratings to confirm the presence of predominant strain, and (b) a series of objective acoustic and psychoacoustic measures previously reported as potentially effective in the characterization of strain.

METHODS

Sustained vowel /ɑ/ productions were elicited from 37 normophonic subjects using habitual (typical) voice and intermediate and maximal levels of strained voice. Strained voice samples were elicited using specific verbal instructions, visual cues, and example voice recordings demonstrating varying levels of strain. Vowel samples were analyzed using measures of sharpness (acum), pitch strength, pitch height, cepstral measures (including cepstral peak prominence [CPP], CPP standard deviation, and CPP fo), measures of spectral tilt (low vs high spectral ratio at 2-kHz and 4-kHz cutoffs), and relative vocal intensity (dB SPL).

RESULTS AND

CONCLUSIONS

Auditory-perceptual analyses showed that productions of intermediate and maximal predominant strain were relatively unencumbered by the covarying effects of breathiness and roughness. The results showed that perceived strain is related to increased sharpness and spectral tilt, increased fo and fo variability, and increased SPL. While minor reductions in CPP and pitch strength may be observed with increasing strain, the presence of relatively strong periodicity and harmonic content in sustained vowel production characterized by predominant strain makes these objective measures weaker indices of increased vocal effort and strain.