Parham Aarabi and Guangji Shi designed a signal enhancement algorithm that employs two microphones and a number of filters. A computer chip analyzes the amount of time it takes sounds to arrive at the two microphones and determines which signals arise from a speaker--and are thus important--and which come from extraneous activity and should be removed. In typical environments there is background noise and reverberations that make it hard to carry on a cell phone conversation, Aarabi says. This system employs two microphones that, just like the two human ears, focus on the speaker's voice and filter out other noises.
The new approach has improved on the performance of alternative speech recognition systems by about 30 percent, the researchers report in the August issue of the journal IEEE Transactions on Systems, Man, and Cybernetics Part B. Other speech recognition systems only reduce the background noise, but this technology also deconstructs other conversations into a slight hum so they don't confuse you, Aarabi explains. The team hopes to have customized chips that can be used to enhance computer voice recognition software ready in the coming months. (The image above shows the computer chip that detects incoming sound signals attached to a Palm Pilot.) Cell phone users will have to wait a few years before the technology is small enough to carry with them, however.