Previous experiments have demonstrated that intelligibility of interrupted speech is dependent upon several parameters including interruption rate, speech-time fraction (STF), ON time, OFF time, and message rate. Our experiments suggest a more parsimonious model that incorporates only ON time and OFF time expressed relative to average word duration. Messages spoken at rates of 186 and 246 wpm were interrupted at rates ranging from 0.42 to 20.0 ips for STFs of 0.67, 0.50, and 0.33. At slow interruption rates, intelligibility was dependent primarily on STF. Intelligibility functions then passed through a minimum located in the vicinity of 1.0–2.5 ips, followed by rapid recovery to nearly 100% intelligibility at 10 ips. The exact location of minimum, however, varied with both message rate and STF. The data can be viewed more simply by a model that incorporates two switching parameters, ON time and OFF time, and temporal characteristics of the message. Equal-intelligibility contours will be described that illustrate the nature of the trading relation between ON time and relative OFF time that is required to maintain a criterion intelligibility level.
CITATION STYLE
Powers, G., & Speaks, C. (1971). Intelligibility of Temporally Interrupted Speech. The Journal of the Acoustical Society of America, 50(1A_Supplement), 130–130. https://doi.org/10.1121/1.1977590
Mendeley helps you to discover research relevant for your work.