As Web-enabled cellular phones become an increasingly popular consumer product, it makes sense to ask whether speech might be a practical way of accessing Web-based information. This study evaluates some of the human factors associated with such a system.

The authors set up an experiment that evaluates user acceptance of speech access to the Web using the technology acceptance model to examine the importance of various factors in the user experience. The characteristics they examine are convenience, fun, accuracy, transaction cost (in this case, the monetary cost of the phone plus the cost of air time to use the system), security, speed, expected time (waiting time and how much time was saved by using the system), and control. In the data analysis, they mathematically extract five components: handling (a combination of availability, waiting time, avoidance of human interaction, and accuracy), accessibility, security, fun, and speed. The authors conclude that handling is the most important factor affecting user reactions to a voice-enabled Web application. Accessibility and fun rank next, followed by security and speed.

This work will be of practical interest to researchers developing voice-enabled systems for the marketplace. The authors also present a methodology that is worth considering when evaluating complex user reactions to new systems.