Acoustic Voice Analysis of Arabic-Speaking Dysphonic Adults Using Praat Software

This study aimed to determine whether Praat-derived acoustic parameters can reliably differentiate abnormal from normal voices. In addition, it examined the relationship between maximum phonation time (MPT), cepstral peak prominence smoothed (CPPS), and the dysphonia severity index (DSI) with perceptual ratings of dysphonia severity.

A total of 178 monolingual Egyptian Arabic-speaking adults aged 21-50 years were recruited and divided into control (n = 100) and case groups (n = 78 with voice disorders). All underwent comprehensive voice assessment, including videolaryngostroboscopy, auditory-perceptual evaluation, durational measurements, acoustic analysis using Praat for sustained vowels (eg, /a/ and /i/), and DSI calculation.

Praat-derived parameters successfully distinguished dysphonic from normal voices. Pathological female groups had higher jitter % and shimmer % than controls. Male groups differed significantly only in fo for both vowels, with no significant differences in jitter %, shimmer % (vowel /a/), or noise-to-harmonic ratio (vowel /i/). Across both sexes and vowels, MPT, CPPS, and DSI values correlated significantly with perceptual dysphonia severity, supporting their clinical utility as objective markers. Logistic regression identified DSI and MPT as significant predictors of dysphonia.

Praat-based acoustic analysis is a reliable tool for distinguishing normal from pathological voices. Perturbation measures (jitter %, shimmer%) appear more reliable in females, while fundamental frequency shifts are more evident in males. Among acoustic and durational measures, DSI and MPT are the most effective for detecting and assessing dysphonia in Arabic-speaking populations.