Validation of the measures of cepstral peak prominence as a measure of dysphonia severity in Spanish-speaking subjects

Abstract

Introduction and objectives: Perceptual rating of overall voice quality and other more specific perceptual dimensions is difficult, as such judgments depend on the listener's subjectivity. Thus, finding objective, valid, and accessible clinical measures to include in comprehensive voice evaluation protocols is a priority. The purposes of this study were to 1) determine the diagnostic accuracy of a single acoustic measure, smoothed cepstral peak prominence (CPPS), to predict voice disorder status from sustained vowels and connected speech samples using the software Praat; 2) to determine the relationship between measures of CPPS and perceptual ratings of vocal quality; and 3) describe the normative values of CPPS.

Method: Measures of CPPS were obtained from connected speech and sustained vowel recordings of 72 Spanish-speaking subjects with voice disorders and 52 nondysphonic Spanish-speaking subjects with no vocal disorders using freely downloadable Praat software. IBM SPSS Statistics software version 23 was used to complete the statistical analyses.

Results: results revealed a 70% sensitivity rate, a specificity rate of 85%. Estimated severity for sustained vowels and connected speech were strongly correlated and significantly associated with listener ratings of dysphonia severity.

Conclusions: A single acoustic measure of CPPS was highly predictive of voice disorder status using Praat software. Clinicians may consider using CPPS to complement clinical voice evaluation and screening protocols.

Keywords: Acoustic measures; Analysis of dysphonia in speech and voice; Análisis de la disfonía en el habla y voz; Disfonía; Dysphonia; Medidas acústicas; Praat; Prominencia del pico cepstral suavizado; Smoothed cepstral peak prominence.