An Empirical Study: Adding Voice Input to a Graphical Editor

This paper presents a user study measuring the time it takes users to create drawings with a graphical editor. Our control group used the mouse as an input device, and our experimental group used a discrete-word voice recognition system in combination with the mouse. Previous studies have measured the speed and accuracy of voice input versus devices such as keyboards and mice; our study measures voice in addition to an existing device. Our primary result is that using voice and mouse input together reduced task completion time by up to fifty-six percent, with an average reduction of more than twentyone percent. We used an existing graphical editor originally designed to receive mouse input; we expect that a system explicitly designed to use parallel input channels could significantly improve our results