Related Experiment Video
Updated: Apr 1, 2026
A Protocol for Comprehensive Assessment of Bulbar Dysfunction in Amyotrophic Lateral Sclerosis ALS
Published on: February 21, 2011
Acoustic Analyses of Prolonged Vowels in Young Adults With Friedreich Ataxia
Cecyle Carson1, Jack Ryalls1, Kaylea Hardin-Hollingsworth1
1Department of Communication Sciences & Disorders, Health and Public Affairs I, University of Central Florida, Orlando, Florida 32816.
Objectives:
Finding measures that track disease progression and determine treatment efficacy is vital for appropriate management in Friedreich ataxia (FA). The purpose of this study was to determine which cepstral- and spectral-based measures extracted from prolonged vowels using Analysis of Dysphonia in Speech and Voice (ADSV) program discriminate between those who have FA and normal voice (NV) peers.
Study Design:
This is a descriptive, prospective study.
Methods:
Initial 2 seconds of prolonged /a/, /i/, and /o/ were analyzed through ADSV from 20 individuals diagnosed with FA and 20 NV individuals. ADSV measures used were cepstral peak prominence (CPP), cepstral peak prominence standard deviation (CPP SD), low/high spectral ratio (L/H ratio), low/high spectral ratio standard deviation (L/H ratio SD), and the Cepstral/Spectral Index of Dysphonia (CSID).
Results:
L/H ratio SD was the only measure where significant differences were found across all vowels between groups. Comparing measures per vowel, the vowel /o/ was significantly different between groups on four of five measures. Discrimination analysis revealed 100% of those in the FA group were classified correctly (sensitivity), whereas 95% of NV members were correctly identified (specificity) when all ADSV measures, with the exception of L/H ratio, were entered.
Conclusions:
Unstable periods of phonation, such as initiations of voice production in vowels, may yield robust acoustic cues in the FA population. ADSV provides measures that, when considered together, have excellent sensitivity and very good specificity. Vowels yielded differing results on ADSV measures; analysis of different vowel types is recommended.
