PORTUGUESE JOURNAL OF SPEECH THERAPY
RPTF | VOLUME 1 | YEAR II
Acoustic analysis of the vowel /a/ according to the degree of dysphonia in adults
Purpose: (1) determine the degree of dysphonia and the mean acoustic values (fundamental frequency, jitter, shimmer, harmonic- to-noise ratio and normalized noise energy) according to gender and degree of dysphonia; (2) determine whether there are sig- nificant differences in mean acoustic values according to gender and degree of dysphonia; (3) determine whether a statistically significant correlation exists between the degree of dysphonia and acoustic parameters; (4) verify if the degree of dysphonia affects the acoustic data; and (5) identify the acoustic predictors of the degree of dysphonia.
Methods: A comparative, corre- lational and cross-sectional predictive study was performed. 214 recordings of clinical files of 54 men (M=33.05 ± 9.09 years) and 160 women (M=36.50 ± 7.57 years) with mild to severe dysphonia were selected. For control purposes, 24 healthy adults, 12 men and 12 women, with a mean age 39.07 ± 6.78 and 35 ± 6.36, respectively were recorded. The degree of dysphonia was reported using an audioperceptual scale. Evaluation and analysis of fundamental frequency, jitter, shimmer, harmonic-to-noise ratio and normalized noise energy of the vowel /a/ were extracted and analyzed.
Results: Only fundamental frequency shows statistically significant differences (p ≤ 0,05) between genders. The acoustic parameters of jitter, shimmer and normalized noise energy increase progressively and the harmonic-to-noise energy decreases progressively overall perceptive assessment of grade, for both sexes, with statistical significance (p≤0,00). There is a significant correlation (p ≤ 0.00) between jitter, shimmer, harmonic noise ratio and normalized noise energy, and the overall severity of voice quality. However, the multinomial analysis considers that only shimmer and normalized noise energy are significant predictive variables of the overall severity of voice quality (p ≤ 0.00).
Conclusions: The acoustic variables shimmer and normalized noise energy have potential to predict the degree of dysphonia.
KEYWORDS: Disfonia, Avaliação audiopercetiva, Avaliação acústica, Adultos