The effect of slow speech on tongue movements & acoustic vowel space distance in speakers with Amyotrophic Lateral Sclerosis
Authors
Advisors
Issue Date
Type
Keywords
Citation
Abstract
With disease progression, the speech of talkers with Amyotrophic Lateral Sclerosis (ALS) becomes increasingly imprecise and unintelligible. Although speaking rate reduction is commonly used as a treatment approach to enhance speech intelligibility in these talkers, its effect on tongue movements and speech acoustics is not well understood. Thus, the purpose of this study was to determine how slow speech affects tongue excursions and speech acoustics in persons with ALS. Further, this study investigated how tongue excursions and speech acoustics differed between persons with ALS and healthy controls. Lastly, this study sought to determine how predictable tongue excursions are based on speech acoustics in persons with ALS. 3D electromagnetic articulography was used to capture tongue movements during speech in five talkers with ALS and five healthy controls. Tongue excursions and the vowel space distance during the production of the vowels /a/ and /i/ in the word "kite" embedded in the sentence "See a kite again" were measured during typical and slow speech. Results showed increased tongue excursions and acoustic vowel space in response to slow speech for persons with ALS and controls; however, the effect was larger in the control group than in the ALS group. These outcomes support the current clinical assumption that slow speech increases tongue excursions and expanded vowel space in persons with speech impairments due to ALS. Although tongue excursions tended to be slightly larger in persons with ALS than controls, vowel space tended to be smaller in persons with ALS than controls. These findings challenged the current assumption that a small vowel space indicate small tongue excursion. The predictability of change in tongue excursions based on change in speech acoustics in response to slow speech was much lower in persons with ALS than controls. Thus, acoustic measures should not be used to infer the underlying speech movements in persons with ALS.