Many neurologic disorders result in a speech impairment (dysarthria). The most important dimension of a dysarthria is its effect on intelligibility. This project examines the intelligibility of the dysarthria associated with five neurologic disorders-- amyotrophic lateral sclerosis, Parkinson's disease, upper or lower motoneuron lesion due to stroke, cerebral palsy, and cerebellar degeneration. Intelligibility reduction in these dysarthrias is studied by a combination of techniques, including single-word intelligibility tests designed to yield acoustic-phonetic profiles of speech error patterns, acoustic signature analyses of selected test words, acoustic analyses of sentence repetition and conversational samples, correction of specific dysarthric errors by digital signal processing, simulations of dysarthria by speech synthesis, and x- ray microbeam studies of speech articulation in selected dysarthric patients. An overarching hypothesis for these studies is that different dysarthrias can be distinctively profiled with respect to 19 acoustic-phonetic features. The combination of research techniques tests for evidence of these features at various levels, including perception, acoustic signal and articulatory pattern. Research also is planned on the development and refinement of an oral sensation test based on static and kinetic cutaneous sensitivity, vibratory sensitivity, and lip and jaw proprioception. The long-term objectives of this research include (1) improved descriptions of dysarthria, (2) refined methods for evaluation of speech and oral sensory function, and (3) new theoretical interpretations of dysarthria based on perceptual, acoustic and articulatory data.
Showing the most recent 10 out of 41 publications
