Breathiness Indices for Classification of Dysarthria Based on Type and Speech Intelligibility
Chandrashekar H M, Veena Karjigi, N. Sreedevi
Abstract
Chandrashekar H M, Veena Karjigi, N. Sreedevi
Abstract
Dysarthria classification based on intelligibility level is useful for speech pathologists for deciding the therapy. However, intelligibility assessment also depends on perceptual attributes like hypernasality, breathiness, slow rate, short pauses etc. These perceptual attributes vary depending on the cause for dysarthria giving rise to different types of dysarthria. In this work, we explore the use of breathiness features for intelligibility assessment of dysarthria and for distinguishing type of dysarthria. Voiced segments from two controlled speakers, two dysarthric speakers with low and mid intelligibility level each from UA database are used in the work. Features were analysed for use in dysarthria intelligibility assessment vs. distinguishing type of dysarthria.
OpenAlex reports 18 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Dysarthria classification based on intelligibility level is useful for speech pathologists for deciding the therapy. However, intelligibility assessment also depends on perceptual attributes like hypernasality, breathiness, slow rate, short pauses etc. These perceptual attributes vary depending on the cause for dysarthria giving rise to different types of dysarthria. In this work, we explore the use of breathiness features for intelligibility assessment of dysarthria and for distinguishing type of dysarthria. Voiced segments from two controlled speakers, two dysarthric speakers with low and mid intelligibility level each from UA database are used in the work. Features were analysed for use in dysarthria intelligibility assessment vs. distinguishing type of dysarthria.
Key concepts: Dysarthria, Intelligibility (philosophy), Breathy voice, Audiology, Speech recognition, Perception, Psychology, Phonation