Computer interfaces are increasingly becoming more visual. As our dependence on graphical interfaces such as "Windows" increases, it is easy to forget that a portion of our population cannot see.

Making computers accessible to those with disabilities is the job of the Adaptive Technology Lab (ATL) here at Harvard. Two years ago, Harvard Media Services acquired responsibility for the ATL and has since provided computer-aid to the handicapped.

According to John Voloudakis, the ATL coordinator, about 20 students consistently use the facilities, up from about nine two years ago. With facilities in the Science Center, Lamont Library and the Aiken Computation Lab, the ATL helps students university-wide with disabilities ranging from visual impairment to dyslexia.

Visually impaired students can benefit from large monitors, talking computers and a Braille printer. For course items such as textbooks which are not computerized, the ATL can scan in material for students and convert it into electronic form. Blind students can then choose to enlarge the text on the monitor or have the computer read the text to them.

Dyslexic students can also benefit from having material read to them. For both dyslexic students and students with carpal tunnel syndrome (repetitive stress injuries of the hands and wrists), voice recognition software is available for writing papers.

Some of this technology has been around for several years, such as the Optical Character Recognition (OCR) software needed to convert scanned texts and handouts into electronic text.

Other technologies, namely voice recognition, are rapidly improving. For example, ATL uses a Window-based package from Dragon Systems called DragonDictate. (A Mac equivalent called Power Secretary, uses the same voice-recognition engine).

After two hours of familiarizing DragonDictate with your voice, the software is remarkably accurate. Speaking in a somewhat choppy voice, pausing briefly after each word, the software will recognize almost everything you say. Voloudakis said that those experienced with the software could reach "typing speeds" of up to 50 words a minute.

The benefits are indisputable. Voloudakis cited one example of a homeless dyslexic man who was introduced to computers and ATL technology. Now he is a writer pursuing an English literature degree at the Extension School.

This technology and the issue of accessibility have had many other interesting side-effects as well. For example, talking computers have decreased the necessity for blind people to learn Braille. Voloudakis stated that the Braille printer at the ATL was last used two years ago and has not been used since, simply because most students take advantage of the available technology.

In fact, the ATL has been negotiating with book publishers to give it textbooks in electronic form to save the time and labor required to scan and convert textbooks itself. But publishers have balked at the idea, primarily because they fear people will widely distribute these books electronically.

A graphical interface is pretty much useless for the blind. How does a computer tell a blind person to move the mouse pointer over an icon and double-click? One way the ATL has dealt with this problem has been to suggest that students use older programs with textual rather than graphical interfaces.

Finally, what does this technology mean for the rest of us? Voice recognition is clearly applicable in many areas. There are many people (Harvard students included) who can't type very fast; voice recognition software can alleviate this problem. Because voice recognition and other technology is getting better and the hardware required is becoming cheaper, we may see more widespread applications of this technology in the future. For now, we can appreciate the opportunities this technology is creating for those who would otherwise be at a great disadvantage.

Eugene E. Kim '96 is former president of the Harvard Computer Society. He may be reached on-line at "eekim@fas.harvard.edu"