Towards Better Intelligibility Testing of Dysarthria
Author | : Michael R. Fraas |
Publisher | : |
Total Pages | : |
Release | : 2003 |
ISBN-10 | : OCLC:52186086 |
ISBN-13 | : |
Rating | : 4/5 ( Downloads) |
Download or read book Towards Better Intelligibility Testing of Dysarthria written by Michael R. Fraas and published by . This book was released on 2003 with total page pages. Available in PDF, EPUB and Kindle. Book excerpt: This investigation attempted to develop a test for assessing intelligibility of native Spanish-speakers with dysarthria. Twenty-five native Spanish-speakers participated in this study. Eleven were diagnosed with Parkinson2s disease and 14 subjects free of neurological disease served as a comparison group. Subjects read a list of Spanish minimal pair phonetic contrasts randomly presented. These phonetic contrasts were prepared by the investigator and have been considered to be sensitive to dysarthric impairment and contribute to speech intelligibility in English speakers with dysarthria. Responses were digitally recorded. A group of 25 native English speaking graduate students in speech-language pathology and 6 native Spanish speaking professionals served as judges in a perceptual task of intelligibility. Judges listened to the digitally recorded speech samples from the Parkinson2s and comparison groups. English speaking listeners recorded responses by transcribing phonetically what they heard. Spanish speaking listeners wrote their responses orthographically. Scores of intelligibility were awarded to each speaker based on the number of correct responses by each listener. The subjects with PD were found to be less intelligible than the comparison group by both the English and Spanish speaking judges. Acoustic analysis of VOT, vowel space, and F2 transitions for each speaker were examined. Significant differences were found between groups for VOT and vowel space. A trend was also exhibited between groups for F2 transitions although this was not found significant at the .05 level. Regression analysis revealed that VOT was a strong predictor of intelligibility, accounting for 57% of the variance.