Purpose: A version of the "smoothed cepstral peak prominence" (i.e., CPPS) has recently been implemented in the program Praat. The present study therefore estimated the correspondence between the original CPPS from the program SpeechTool and Praat's version of the CPPS. Because the CPPS is the main factor in the multivariate Acoustic Voice Quality Index (AVQI), this study also investigated the proportional relationship between the AVQI with the original and the second version of the CPPS.
Study design: Comparative cohort study.
Methods: Clinical recordings of sustained vowel phonation and continuous speech from 289 subjects with various voice disorders were analyzed with the two versions of the CPPS and the AVQI. Pearson correlation coefficients and coefficients of determination were calculated between both CPPS-methods and between both AVQI-methods.
Results: Quasi-perfect correlations and coefficients of determination approaching hundred percent were found.
Conclusions: The findings of this study demonstrate that the outcomes of the two CPPS-methods and the two AVQI-methods are highly comparable, increasing the clinical feasibility of both methods as measures of dysphonia severity.
Keywords: Accuracy; Acoustic voice quality index; Feasibility; Praat; Smoothed cepstral peak prominence; SpeechTool.
Copyright © 2015 The Voice Foundation. Published by Elsevier Inc. All rights reserved.