Every year tens of thousands lose their larynx – their voice – to infections, injury, or cancer. Various forms of larynx substitutes are in use; the common approach seen in movies and television is holding a resonating box to the throat (or a hole in the throat). All of these approaches sound very peculiar. A team of researchers from South Africa are demonstrating a new approach, which combines a palatometer with a computerized voice synthesizer.
“All of the currently available devices produce such bad sound–it either sounds robotic or has a gruff speaking voice,” says Megan Russell, a PhD candidate at the University of the Witwatersrand in Johannesburg, South Africa. “We felt the tech was there for an artificial synthesized voice solution.”
The system uses a palatometer: a device that looks much like an orthodontic plate and is normally used for speech therapy. The device, made by CompleteSpeech of Orem, UT, tracks contact between the tongue and palate using 118 embedded touch sensors. The software for the artificial larynx was written by Russell and colleagues at the University of the Witwatersrand.
To use the device, a person puts the palatometer in her mouth and mouths words normally. The system tries to translate those mouth movements into words before reproducing them on a small sound synthesizer, perhaps tucked into a shirt pocket.
So far, Russell has trained the system to recognize 50 common English words by saying each word multiple times with the palatometer in her mouth. The information can be represented on a binary space-time graph and put into a database. Each time the user speaks, the contact patterns are compared against the database to identify the correct word.
[Source: Technology Review]
The palatometer approach has several advantages in that it requires no surgery and can be removed or replaced at will. On the other hand, the difficulty of translating palatal and other mouth movements is very much akin to those of voice recognition. People are different in the way they form their words. Sometimes very subtle changes in mouth configuration can have profound effect on the pronunciation of words and letters. Consequently the South African team has spent much of its time in a continual process of refining the interpretive software for the palatometer – attempting to widen the recognizable vocabulary and improve the rate of accuracy.
It’s fair to say that some kind of ‘mouth reader’ is in the works, whether it is the palatometer approach or something else. There is still much work to be done to create accurate sensor devices that can stay in the mouth for long periods of time; that communicate wirelessly with computer devices; and that are interpreted by fast, accurate, and expressive voice reproduction software.