Spectral, cepstral, and short-term acoustic measures in women with and without laryngeal and vocal alterations.

Abstract

Purpose: To compare acoustic voice measures of women with and without laryngeal and vocal alterations.

Methods: Cross-sectional, analytical, observational study with 69 female teachers, divided into four groups: 21 with laryngeal alterations (LAG), 48 without alterations (NAG), 45 with vocal deviations (VDG), and 24 without deviations (NDG). The groups were defined by otolaryngological evaluation (videolaryngoscopy or videolaryngostroboscopy) and by auditory-perceptual evaluation of the grade of hoarseness. Participants who smoked, were pregnant, menstruating, or had allergies or the flu were excluded. Voice and speech samples (3 central seconds of the vowel [a] and counting from 1 to 11) were recorded and edited using Praat. The acoustic measures jitter, shimmer, harmonic-to-noise ratio (HNR), glottal-to-noise excitation (GNE), smoothed cepstral peak prominence (CPPS), spectral tilt, high-frequency noise (HF noise), and the difference between the first and second harmonics (H1-H2) were extracted from the VOXplot. Statistical analysis was performed using the t-test, Mann-Whitney test, and Hedges' g test.

Results: Women with laryngeal and vocal alterations had lower GNE, HNR, CPPS, H1-H2, and HF noise values and higher jitter, shimmer, and spectral tilt values. CPPS, spectral tilt (vowel), and H1-H2 (vowel and speech) were not different between the groups with and without laryngeal alterations. CPPS (vowel) and spectral tilt (vowel and speech) were likewise not different between the groups with and without vocal deviation.

Conclusion: The acoustic measures responded similarly to the presence of laryngeal and vocal alterations, reinforcing the combined influence of structural and functional factors on the quality of the vocal signal.