Intelligibility of sequentially presented speech signals.

Abstract

In some operational environments, listeners are required to monitor several channels of simultaneous time‐critical communication. There are two possible ways to reduce the transmission time for multiple speech messages: (1) present information from two or more speech channels simultaneously or (2) time compress the competing speech messages and present them sequentially. Previous studies have shown that the intelligibility of a single speech message is fairly impervious to high degrees of time compression. However, it is not clear if intelligibility is preserved when several such highly compressed messages are presented sequentially. The current experiment investigated speech intelligibility when two, three, or four time‐compressed speech signals were presented sequentially. Three results were apparent: (1) speech intelligibility was near ceiling when time‐compressed signals were presented sequentially and listeners were required to identify and respond to only one of the signals in the sequence; (2) speech intelligibility decreased dramatically when the listeners were required to respond to all the messages in the sequence, particularly at high compression rates; and (3) overall performance in the multiple‐message identification task depended primarily on the total time interval between the onset of each new message, and not on the degree of time compression of the speech message contained in that interval.