Few stories epitomize the transforming power of assistive technology like the rise of Neil MacGregor. Diagnosed with a learning disability as a child, he was illiterate in middle school, and still reading at a seventh-grade level when the University of Guelph rejected him.
MacGregor got in to Guelph only as part of a national study exploring what assistance LD college students would need to succeed. "A lot of answers emerged," said MacGregor. "But effective use of assistive technology was the biggest."
Computer technology and auditory learning (listening while reading text brought Shakespeare to life) propelled MacGregor to the top of his class. He earned a B.A. in Philosophy and graduate scholarship from the same school that had him once turned him away. MacGregor turned it down to train other LD students (and later, teachers) how to use technology to enlarge learning.
MacGregor is now Vice President of Learner Development for GoQ Software, whose flagship product, wordQ+speakQ, is among the most popular writing programs for LD students, used by millions worldwide. It's installed on every school computer in Newfoundland and Ontario and district-wide in US cities, including all 131 schools in Albuquerque, New Mexico.
wordQ+speakQ's killer app is how its trio of technologies (word prediction, voice recognition, and text-to-speech) augments "scaffolding," the process through which students expand current skills to support new knowledge.
In this interview, conducted by phone on April 21, 2011, MacGregor talks about the role of assistive technology in education, especially how voice recognition and text-to-speech strengthen, but never replace, crucial cognitive functions all students must develop.What Types of Students do Your Products Help?
Our products were originally designed for people with dyslexia and others who struggle to express themselves the way they want to when they write. We found the number of people who struggle with writing to be significantly higher than those diagnosed with a specific learning disability.What Assistive Technologies do the Use?
Our product integrates word prediction, voice recognition, and text-to-speech. All three have been around a while; all are available elsewhere. What's unique about wordQ+speakQ is how they layer the tools so they can be used simultaneously without adding to a student's workload. The design enables individuals to type the words they know, say the words they don't, and select others that the program predicts or presents. It's not automated corrections. The goal is not to fix things, but to give students assistance when they need it, without constantly tripping over it or relying on it too much.How Does Technology Bolster Students' Current Knowledge?
Students often have no trouble spelling small words, but struggle with big ones. Speech recognition is just the opposite: short, simple words are the hardest to get right because they're small and sound similar to many others. "Ball" is a good example. Longer and more challenging words limit possibilities for confusion and are easier for the software to recognize. People often imagine using speech recognition to dictate entire paragraphs, but it's designed to enable users to develop ebb and flow among what they type, and what they ask the machine to help them with.What Educational Benefit Does Text-to-Speech Provide?
Spoken feedback gets the least lip service but is probably the most effective component in our program. People don't think about "listening" to their work, but we often identify mistakes more easily by sound than by sight. Think how clearly you hear a person who is learning English make pronunciation or grammatical errors. We speak words long before we learn to write them. Our technology integrates the visual image of words onscreen with the auditory modality of how things are supposed to sound; sensitizing students so the computer's speech corresponds to what they want to hear. We enable students to use a resident skill (listening) to enlarge and reinforce learning.Do Consumers Like Synthetic Speech?
Reactions vary. Many don't want spoken feedback, but a growing segment does, so we offer it. wordQ does word prediction and speech; speakQ does voice recognition. They sell separately, but integrate when installed together. Our speech program comes from A Capella, which makes voices that sound amazingly real-they're developing one that sounds like a radio DJ-and enable users to control inflection, intonation, speed, and pitch.Why Does Voice Quality Vary Among Accessible Products?
Quality is crucial in assistive devices aimed at consumers; less critical in others, e.g. screen readers for the blind. JAWS for Windows, for example, speaks everything onscreen, every keystroke and action. Description and presentation order are paramount; voice quality is secondary. For LD students and the general public, it's just the opposite: they only want it to read what they tell it to read and they want it to sound amazing. I trained over 600 LD students over three years and saw what happens when you cheap out on voice: the kids tune it out, and you're gone. With adults, this happens even faster.How Does wordQ Compare to Integrated Readers Such as Apple VoiceOver?
VoiceOver works well with the new operating system, but doesn't enable you to vary the speed. This is crucial to those of us who read using technology. You read a legal document at the same speed as an email from your mom. One problem I have with Apple (and I've got an iPhone and iPad) is not having any control over the reading speed.