Effects of Speech Mode on Anticipatory Coarticulation in Dysarthria

K Tjaden, G Wilding, D Rivera, A Kozak, Y Yu


The current study compared the effects of rate reduction and increased vocal loudness on anticipatory coarticulation for speakers with Multiple Sclerosis (MS), speakers with Parkinson disease (PD), and healthy controls, as inferred from the acoustic signal.  Speakers produced target words in a carrier phrase while being audio recorded in Habitual, Loud and Slow speaking conditions.  First moment coefficients for consonants and vowel F2 frequencies were used to infer coarticulation within syllables and between syllables.  The finding that there were few group differences in coarticulation supports a growing number of studies suggesting that the organizational coherence of phonetic events is generally preserved in mild to moderate dysarthria. Speech mode effects (e.g., increased loudness, reduced rate) on coarticulation were present for only a few phonetic contexts, but when present, tended to be similar for healthy and disordered speakers.  These results suggest that coarticulatory adjustments likely do not strongly contribute to improvements in intelligibility and precision that have been reported when speakers with dysarthria reduce rate or increase vocal intensity.  In addition, the finding that increased loudness and reduced speech rate did not strongly or differentially affect acoustic measures of coarticulation for disordered speakers needs to be reconciled with studies suggesting different neuromuscular controls strategies for Loud and Slow speech produced by speakers with dysarthria.