Augmenting Comprehension of Speech in Noise with a Facial Avatar and Its Effect on Performance

摘要

Military operations often occur in noisy environments, which can interfere with effective verbal communication. Previous studies have established the effectiveness of allowing a listener to see the speaker's mouth. This study examined the efficacy of incorporating a computer-animated facial avatar into a visual display in order to improve the comprehension of speech in noisy environments, while performing concurrent tasks. It also examined the effect of the avatar on the performance of concurrent auditory and visual tasks. Twenty volunteers participated in an experiment measuring verbal comprehension, concurrent task performance and gaze dwell times while auditory, verbal and visual tasks were being performed under noisy conditions. The results indicated that the simple presence of the facial avatar did not significantly improve verbal comprehension while performing concurrent tasks. However, the facial avatar significantly improved verbal comprehension when the tasks being completed concurrently were more difficult and/or auditory-type tasks. The participants' performance for the concurrent tasks was not significantly affected by the presence of the facial avatar. The incorporation of computer-animated facial avatars into visual displays has the potential to improve verbal comprehension in noisy environments, depending on the nature of the concurrent task.