Off-campus UNL users: To download campus access dissertations, please use the following link to log into our proxy server with your NU ID and password. When you are done browsing please remember to return to this page and log out.
Non-UNL users: Please talk to your librarian about requesting this dissertation through interlibrary loan.
Duration and variability in dysarthric speakers with traumatic brain injury
Abstract
Dysarthria of speech can occur in many acquired and congenital neurological conditions and can impact the articulatory, respiratory, phonatory, laryngeal and/or velopharyngeal subsystems of speech. Studies of dysarthria describe the differences that occur in speech performance due to different types and severities of dysarthria. The conditions that cause dysarthria are often accompanied by severe physical impairments that make standard access to computer technology inefficient. Many individuals with physical disabilities use a wide range of assistive technology options to access computers. However, if a co-occurring dysarthria is present, speech recognition (SR) technology is not an option. Studies of SR use by people with dysarthria describe the limited success many have with such technology. Limited investigations exist examining the impact of specific acoustic characteristics of dysarthric speech on SR technology. This study investigated the durations of single words and sound types with acoustic analysis as well as the variability of word durations of ten participants with dysarthria due to Traumatic Brain Injury (TBI) and ten control participants. The study also examined the relationships between word intelligibility and word duration and word intelligibility and variability for the participants with TBI. Results showed statistically significant differences on word and sound type durations between the dysarthric and control participants. Specifically, a pattern of doubling or near-doubling of durations was consistent across word and sound type durations. Extraneous features were identified from the acoustic signals of the dysarthric speakers and included pre-vocalizations, insertions, omissions, substitutions, and voicing of voiceless consonants. When controlling for these features, the word and sound type durations remained significantly greater for the dysarthric compared to the control participants. Differences existed in standard deviations but not the coefficient of variability between dysarthric and control speakers. There was no correlation between word intelligibility and word duration or word intelligibility and variability. Implications for SR technology development based on the study's findings were discussed.
Subject Area
Physical therapy|Speech therapy
Recommended Citation
Fager, Susan K, "Duration and variability in dysarthric speakers with traumatic brain injury" (2008). ETD collection for University of Nebraska-Lincoln. AAI3331384.
https://digitalcommons.unl.edu/dissertations/AAI3331384