Kyuhan's Assignment 13

by Dale-Marie Wilson, Auburn University, 2006

Reviewed by Kyuhan Koh, University of Colorado at Boulder, Dec 2007

In spite of the great improvement of technology, the most popular mediums are still paper and online documentation for technical assistance. As technology progresses, there is a need to address the level of performance of the current mediums used for technical communications.

This paper addressed a new medium to users for technical assistance. In other words, it introduced an innovative way for a manual. With this dissertation, Dr. Wilson provided a prototype of iTech, a conversational technical assistant developed using the technical communications from a vi manual published by O’Reilly and showed the level of performance of three different mediums such as paper based, online based, and iTech based assistance.

iTech is an interactive agent system with natural language processing. An animated agent in iTech interface accepts user’s queries as spoken language, and if the question is matched, the solution is displayed.

The basic technical features of iTech include:

• The built-in speech recognition engine, Microsoft English ASR Version 5 Engine, recognizes the user’s question and passes the recognized speech to the browser environment of the page where the Speech Application Language Tags (SALT) is hosted.

• The graphical user interface (GUI) consists of two frames: the Navigation frame and the Content frame The Navigation frame consists of the animated agent and the Speech Application Language Tags (SALT). JavaScript was used to control iTech’s speech. SALT is then used to enable iTech’s hearing.

Unfortunately, there is no significant technical progress in this paper as I think. However, in the point of user-centered design, I would like to say it is a step to go further for user-friendly design to provide technical support. It was my first time to see the interactive manual with speech recognition, and I liked her idea very well. This kind of approach for technical assistance would be beneficial for senior citizens and young children who are not familiar to use online technical assistance.

The idea of Dr. Wilson sounds promising and innovative, but the experimental result was a little bit disappointing. Although iTech showed the fastest average search time, almost four times faster than other mediums, it failed to give good impression for user satisfaction, average task completion time, and reading time representing the time from the appearance of the solution on the monitor to the time the participant touched the keyboard. Besides the accuracy rate of current speech recognition engine, I could not say the performance of iTech was greatly better than current mediums. I believe that these failures came from the way of delivering the content. The methodology is combined with speech recognition, agent system, and natural language processing, but the content is still based on the book. To be effective user interface design, the design itself is important, but revising the content to utilize the design fully is more important.