Select your localized edition:

Close ×

More Ways to Connect

Discover one of our 28 local entrepreneurial communities »

Be the first to know as we launch in new countries and markets around the globe.

Interested in bringing MIT Technology Review to your local market?

MIT Technology ReviewMIT Technology Review - logo

 

Unsupported browser: Your browser does not meet modern web standards. See how it scores »

The team hopes to eliminate the palatometer’s ribbon cables, which run out of the user’s mouth, and instead create a system in which data is transmitted wirelessly from the palatometer to a speech synthesizer. The group also needs to improve upon the predictive analysis system and expand the database of words.

Russell’s team will also need to test many more subjects, including those without a larynx, before the device could become widely available.

“We also intend to implement a degree of user-controlled modulation of pitch and volume in order to achieve a more natural sound,” says David Rubin, an adjunct professor of electrical and information engineering at the University of the Witwatersrand and one of Russell’s advisors on the project. “For example, we intend the user to be able to achieve the typical upward inflection in the voice when asking a question as compared to making a statement.”

It is also important for the team to improve on the processing speed of the device. It now takes more than a second for the system to identify and play back the correct word. “Ultimately this time needs to be reduced to below 0.3 seconds in order for it to appear to observers that the person’s lips and voice are synchronized,” Russell says.

Researchers at the University of Hull in the U.K. are exploring a different approach to the problem. They place magnets in the mouth and these to measure changes in the magnetic field around the mouth that correspond to movement. Currently, the Hull researchers use surgical glue to affix six magnets to the lips, throat, and tongue. Ultimately, the magnets would be implanted. “We are still working out where is the best place to put them,” says James Gilbert, a senior lecturer in engineering at Hull.

Like the Witwatersrand team, the Hull group hopes to eliminate much of the wiring and expand its dataset. Currently, the Hull system can only identify 10 words, and the accuracy can range from anywhere between 70 and 100 percent.

However, Gilbert questions whether all words could ever be identified using a palatometer. “The analysis methods seem reasonable,” he says, but adds that “for some words, there’s very limited contact with the palate.”

The Witwatersrand team believes its design is superior because it doesn’t require surgical implants. “In the event that the overall idea proves to be feasible, we hope to be able to enter into agreements with the companies producing existing systems to enable us to move forward with this approach to an artificial larynx,” Rubin says.

0 comments about this story. Start the discussion »

Credit: Jaren Wilke/Megan Russell, The University of the Witwatersrand

Tagged: Biomedicine, robotics, data, voice, prosthetics, device

Reprints and Permissions | Send feedback to the editor

From the Archives

Close

Introducing MIT Technology Review Insider.

Already a Magazine subscriber?

You're automatically an Insider. It's easy to activate or upgrade your account.

Activate Your Account

Become an Insider

It's the new way to subscribe. Get even more of the tech news, research, and discoveries you crave.

Sign Up

Learn More

Find out why MIT Technology Review Insider is for you and explore your options.

Show Me