FROM BOBST LIBRARY

Voice-Input Technology
and Repetitive-Stress Disorder

by Tom McNulty

[Ed: Links to web pages and/or e-mail addresses which have become inactive since the publication of this article have been enclosed in curly brackets { }. Replacement links have been provided where possible.]

Individuals with repetitive-stress disorders, including carpal-tunnel syndrome, represent one of the fastest-growing disability groups in the nation. A 1994 New York Times article states that over 282,000 cases have been reported nationally, but most ergonomics experts agree that the number is, in fact, much larger. This national trend is reflected in the growing number of students who seek assistance from NYU's Henry and Lucy Moses Center for Students with Disabilities. According to Jack Gentul, director of the center, "increasing numbers of individuals with repetitive stress injuries are attending the university; the number of para- and quadriplegics has also jumped significantly in the past two years." For this population, voice-recognition technology holds the key to independent access to computers, and this ability is essential for students to succeed at the university.

Voice-Input Technology

A few years ago, Bobst Library introduced a personal computer with voice input that was intended for users with disabilities. Requests for training have grown steadily over the last two years, but few users have been able to master this complex and fairly primitive system. This fall, the library is introducing a new voice-input system for the university's growing population of library users who have difficulty with regular computer keyboards. The library's first voice-input system had a vocabulary of only a few thousand words. By contrast, the new system, Dragon Dictate, provides a base vocabulary of 30,000 words.

Here's how it works: The Dragon Dictate user enters text and other computer commands into a high-quality, "hands-free" microphone. Dragon's online tutorial allows new users to learn about and customize the program simultaneously. Throughout the tutorial, users are asked to repeat certain words and phrases; Dragon Dictate compares the individual's unique speech patterns against word models stored in the program; powerful statistical and mathematical algorithms then analyze the phonetic features of the spoken word to determine if it matches one of the words stored in the program.

Dragon Dictate continues to refine its understanding of the individual user's voice characteristics through continued use. The program also has some built-in features, including "smart number handling." We generally pronounce the words two, too, and to in exactly the same way. Dragon's powerful software contains rules of English usage that predict the correct form based upon context and probability. If, for whatever reason, the wrong word does appear on the screen, the user can simply say "Oops," and a numbered list of the ten closest possibilities appears on the screen; from this list, the user simply states the number of the correct option and that word is substituted in the document.

For More Information

If you would like more information about adaptive technologies and services provided by NYU's Henry and Lucy Moses Center for Students with Disabilities, contact Yolanda Cacciolo at 998-4976; the office is on the 7th floor of Loeb Student Center.

To make an appointment for a demonstration of Dragon Dictate or Bobst Library's other special equipment for persons with visual, auditory, or motor impairments, please contact me at 998-2519. [ C ]

Tom McNulty is Reference Librarian for the Fine Arts and Coordinator of Bobst Library's Services for Students with Disabilities.
{mcnulty@elmer1.bobst.nyu.edu} Replacement address: tom.mcnulty@nyu.edu

Posted 26 October 1995. Revised 30 January 2004.