Functioning as miniature amplifiers in the ear or outside it, hearing aids and cochlear implants not only increase the volume of hard-to-hear voices, but they ramp up background noise as well. This makes it hard for the hard of hearing to make sense of what’s being said when they are in a busy work environment or at a café, says the professor.
In these situations, “Hearing-impaired people have a real problem understanding speech,” she explains. “Their devices may be useful in a quiet room, but once the background noise levels ramp up, the devices become less useful. Our algorithm helps filter out irrelevant noise so they can better understand the voices of their friends and family.”
Drowning out the noise
Using a cochlear model that she developed, the new patented invention is easy to install on existing implants and hearing aids, requiring companies to insert it into their operating software. The university’s tech transfer company Ramot is currently in talks with several manufacturers.
How does it work? Using math, says Furst-Yust: “We’ve developed a mathematical model of the ear that shows how speech recognition works. The math is complicated, but basically we’re cleaning auditory information before it goes to the brain. We get rid of some of the information – the background noise – so that the hearing-impaired have an easier time ‘filling in’ missing information that their ears can’t give them,” she says.
First developed for cell phones to make it easier to hear voices on the receiver’s end, the researchers found it distorted the human voice too much, making Clearcall uncomfortable for most people to use. That’s when Furst-Yust started applying her tools to hearing aids.
Fit for an iPod?
For people who are born deaf or who are used to listening through aids, understanding the way Clearcall distorts the human voice is something they can more readily accept. “It takes some getting used to,” she says, “but people who have been wearing hearing aids all their lives have no problem getting the most from Clearcall. And we can train the newly hearing impaired in a quick introductory session.”
Using the brain’s own sound recognition centers, Clearcall helps the wearer separate voices from background noise. With continued use and a short training period, the Clearcall software can improve the clarity of voices from 30 to 50%, says Furst-Yust. She is currently putting her results into a scientific paper.
Meanwhile, Furst-Yust continues to refine her algorithm for future applications. She imagines that this base technology will one day have a number of specific functions beyond servicing the hearing impaired. She doesn’t rule out a noise-canceling device to dampen the sound of loud iPod music coming from people’s earphones on the bus and subway.