Independent Perspective News

LONDON, England (CNN) -- When it comes to entering computer data without a keyboard, the eyes have it.

Replacing a keyboard or mouse, eye-scanning cameras mounted on computers have become necessary tools for people without limbs or those affected with paralysis. The camera tracks the movement of the eye, allowing users to "type" on a virtual keyboard as they look at the screen.

And now, researchers from the Cavendish Laboratory at Cambridge University say they have developed software that replaces the standard QWERTY keyboard layout with one that is nearly twice as efficient, more accurate and easier on the eyes.

A report on the software appears in Wednesday's edition of the journal Nature.

Called Dasher, the prototype program taps into the natural gaze of the eye and makes predictable words and phrases simpler to write, said David MacKay, project coordinator and physics professor from Cambridge University.

MacKay compared Dasher to controlling a vehicle with just visual cues. "Something that everyone does is look where they're going, like driving a car," said MacKay. "To write with Dasher, you just look at the screen for what you need and steer the cursor into place."

Looking for letters

Any off-the-shelf camera capable of scanning eye movement can be used with Dasher, though the person must sit fairly still during the interaction.

The letters of the alphabet appear in a single column on the right of the screen, with an underscore symbol to represent a space. Each letter is framed by a colored box.

As the user looks at a particular letter on the right side of the screen and drags it to the left with their eye, another sub-alphabet column begins to emerge inside the box on the right-hand side, along with more letters framed in colored boxes.

Dasher is designed to anticipate which letter will be needed, so although the successive sub-alphabet columns are initially very small, the letters or combination of letters that simultaneously appear are most likely to be used next in that sequence.

For example, if a person starts with the letter "h," the language models in Dasher will bring up "a," "e," "i," "o" and "u" in the sub-alphabet box, along with a few other possible combinations like "ello" to form the word "hello." Each box has a complete alphabet within it, though the first letters to appear have the highest probability of usage.

The letters are then placed together to form a sentence on the left side of the screen, and the process speeds up over time as Dasher learns from each typing experience and saved document.

An eye to the future

Researchers say people will be able to write up to 25 words per minute with Dasher compared to on-screen keyboards, which they say average about 15 words per minute.

With a bit of practice, MacKay said, Dasher offers an easier and more satisfying way for disabled people to communicate, providing them with better tools to write e-mail or create word processor files.

Dasher also can be used with a mouse or tracking ball instead of the eye-scanning process, and MacKay said it might lure some people away from the traditional QWERTY keyboard.

It also has potential for mobile applications, such as a personal digital assistant, with which people could use a stylus pen on the tiny screen to write information.

MacKay said Dasher could work in most languages.

He added that the coding of the software will be made available to the open-source community for development of programs for various computer systems.