The effects of perceptual learning on capacity demands for recognizing synthetic speech
Lisa Lee, Howard C. Nusbaum · The Journal of the Acoustical Society of America · 1989
Previous studies have demonstrated that recognition of synthetic speech improves with moderate amounts of training. Three experiments were performed to determine whether the effects of training are due to changes in the attentional capacity required to process the speech. Attentional capacity available for recognizing speech was manipulated using a recall task with different length lists of visually presented numbers. In two experiments, before and after training, the primary task required subjects either to recall lists of words or to perform speeded word recognition for synthetic speech. Whereas training improved performance with synthetic speech, decreasing available capacity impaired performance. In a third experiment, the intelligibility of the speech was manipulated by using speech from different text-to-speech systems. The results of these experiments suggest that neither training nor intelligibility affects the amount of attentional capacity required to recognize synthetic speech. Rather, as perceptual learning occurs, listeners begin to allocate their attentional resources more efficiently to the processing of synthetic speech. [Work supported by AFOSR and BRSG.]