IEEE TRANSACTIONS ON SPEECH AND AUDIO PROCESSING, VOL. 10, NO. 8, NOVEMBER 2002
Concept design for (a) MiPad’s first card and (b) MiPad’s calendar card.
TABLE III THREE EXAMPLES SHOWING BENEFITS TO COMBINE SPEECH AND PEN FOR MIPAD USER INTERFACE
C. User Study Results
Our ultimate goal is to make MiPad produce real value to users. It is necessary to have a rigorous evaluation to measure the usability of the prototype. Our major concerns are:
“Is the task completion time much better?” and “Is it easier to get the job done?”
For our user studies, we set out to assess the performance of the current version of MiPad (with PIM features only) in terms of task-completion time, text throughput, and user satisfaction. In this evaluation, computer-savvy participants who had little experience with PDAs or speech recognition software used the partially implemented MiPad prototype. The tasks we evaluated include creating a new appointment and creating a new email. Each participant completed half the tasks using the tap and talk interface and half the tasks using the regular pen-only the iPad interface. The ordering of tap and talk and pen-only tasks was random but statistically balanced.
1) Is Task Completion Time Much Better?: Twenty subjects were included in the experiment to evaluate the tasks of creating a new email, and creating a new appointment. Task order was randomized. We alternated tasks for different user groups using either pen-only or Tap & Talk interfaces. The text throughput is calculated during e-mail paragraph transcription tasks. On av- erage it took the participants 50 s to create a new appointment with the Tap & Talk interface and 70 s with the pen-only inter-
face. This result is statistically significant with
. Time savings were about 30%. For transcribing an email it took 2 min and 10 s with Tap & Talk and 4 min and 21 s with pen-only. This difference is also statistically significant,
. These time savings were about 50%.
Error correction for the Tap & Talk interface remains as one of the most unsatisfactory features. In our user studies, calendar access time using the Tap & Talk methods is about the same as pen-only methods, which suggests that pen-based interaction is
suitable for simple tasks.