SPEECH RECONNAISSANCE: Wearable Computers
Our goal with this column is to spot emerging trends in speech recognition, to be a little ahead of the curve in an extremely fast paced industry. For that reason we could not resist a trip to the First Annual Symposium on Wearable Computers held in October in Cambridge, Mass. near the MIT campus. Once there, it was hard not to notice the similarity between where wearable computers are now and the position speech technology was in just a short time ago. Some of the press attention for the wearable computer show, noting coincidence of the Oct. 12 date, compared it to Columbus' discovery of America. Using that analogy, speech may be closer to being in its Industrial Revolution phase. Wearable computers are to a large extent made possible by the advent of speech technology. Both benefit from the continued advances in semi-conductor technology which allow for high performance microprocessors requiring less power and more space. "Decades of research in computer science have provided the technology for hands-off computing using speech and gesturing for input," said Dr. Len Bass of Carnegie Mellon University. "Combined with mobile communication technology, it is possible for users to access information anywhere." While speech technology veterans would no doubt recognize some of the uses of wearable computers such as command and control applications for assembly lines and warehouses, there were also other applications such as virtual vision and touch interfaces. In several cases, "wearable" did not mean a headset and belt but rather referred to "smart fabrics" - microprocessors which can be woven into the fabric of wash and wear clothes. The speech and wearable worlds came together in the form of ISAAC (Integrated Speech Activated Application Control), a prototype developed at Carnegie Mellon in which a wireless microphone transmits analog speech to a speech recognition system on a base computer. An ISAAC user could control office systems, presentations, and e-mail from anywhere in the building. Although our view has always been that 25th century technology will get here in the 25th century, the Star Trek model suddenly seems much closer than we had realized.
Do you have news about a speech technology product, story, or idea?
Contact Brian Lewis, Editor: Speech Technology Magazine, CI Publishing, 43 Danbury Road, Wilton, CT 06897. E-mail: Speechmag@aol.com