A former computer games programmer claims to have invented a "revolutionary" voice recognition system.

Simon Nicol, co-owner of Hove-based At Works Computers, says he has designed a "listen and repeat" software package capable of recognising words, regardless of pitch, tone or accent.

He claims the technology will pave the way for a "Star Trek" scenario where humans and computers communicate on an extraordinarily sophisticated level.

It could also have huge implications for the way visually impaired people use computers.

Simon believes the formula will completely change how sound is recorded, manipulated and reproduced, making the graphic equaliser redundant.

He has worked on the project for seven years after giving up computer game programming to do something that would "stand the test of time".

He enlisted the help of friend Justin Scarlett, a research scientist, who has designed a microphone to separate frequencies more accurately, producing a superior sound.

Current voice recognition systems work by using five to ten minutes' dictation to build up a "personal voice model" based on how the user is likely to pronounce words.

Simon and Justin claim to have developed a more sophisticated programme which focuses on the intrinsic shape of words rather than someone's particular way of speaking.

Now the pair are looking for backers and a manufacturer to help them develop a working prototype they can demonstrate to software publishers.

Simon, 37, said: "The technology that is out there now is fundamentally flawed because it is all about condensing sound and producing an average.

"At the moment, the process is hopelessly laborious - you have to repeat phrases three or more times so the computer can work out, store and respond to an 'average'.

"The problem is, unless you say a word almost exactly the same way as you had done previously, it often spews out gibberish.

"I asked myself what was uniquely special about words and their shape that a computer might be able to recognise, so that it wouldn't matter if it was said in a Scottish accent or a cockney accent."