Influence of visual information on the intelligibility of dysarthric speech

Abstract

Purpose. The purpose of this study was to examine the influence of visual information on the intelligibility of dysarthric speech. The two research questions posed by this study were: (1) Does the presentation mode (auditory-only versus auditory-visual) influence the intelligibility of a homogeneous group of speakers with dysarthria? and (2) Does the experience of the listener (experienced versus inexperienced with dysarthric speech) influence the intelligibility scores of these speakers? Background. Investigations of speakers with hearing impairment and laryngectomy have indicated that intelligibility scores are higher in an auditory-visual mode compared to an auditory-only mode of presentation. Studies of speakers with dysarthria have resulted in mixed findings. Methodological issues such as heterogeneity of speaker groups and factors related to the stimuli may have contributed to these mixed findings. Method. Eight speakers with dysarthria related to Parkinson disease were audio and video tape-recorded reading sentences. Movie files were created in which an auditory-only condition containing the speaker's voice but no visual image of the speaker and an auditory-visual condition containing the speaker's voice and a view of his/her face. Two groups of listeners (experienced and inexperienced with dysarthric speech) completed listening sessions in which they listened to (auditory-only) and watched and listened to (auditory-visual) the movies and transcribed what they heard each speaker say. Results. Although auditory-visual scores were significantly higher than auditory-only intelligibility scores, the difference between these scores was influenced by the order in which the two conditions were presented. A speaker effect was found across presentation modes, with less intelligible speakers demonstrating greater benefit from the inclusion of visual information. No statistically significant difference was found between the two listener groups in this study. Conclusions. These findings suggest that clinicians should include assessment of both auditory-only and auditory-visual intelligibility measures in speakers with Parkinson disease. Management of intelligibility impairment in these individuals should consider whether visual information is beneficial to listeners.