Identification of an acceptable mixture of key and speech inputs in bimodal interfaces

This study was designed to determine the acceptable mixture level of key and speech inputs in a bimodal interface in which users were permitted to use both key and speech input systems. The mixture level was a controlled experimental variable. Five mixture levels were used: 0%, 25%, 50%, 75%, and 100%. The experimental task was to edit a document. The 0% mixture level meant that the participant performed an editing task using only the keyboard. At the 100% mixture level, the participant performed an editing task using only speech input. After each experimental condition had been tested, the participants' mental workload was also measured using the National Aeronautics and Space Administration-Task Load Index. The results suggest that the mixture level between 25% and 50% was acceptable in view of both the completion time and workload.