They look the same, have identical genetic make-ups and even sound the same.

In fact identical twins Fiona and Lindsey McGregor, pictured, sound so similar even their own parents, siblings and boyfriends cannot tell them apart on the phone.

But having uttered just a few words each Jonathan Smith can now identify them within seconds.

His technology firm, the Shipley-based ICR Group, has developed a voice recognition system which could revolutionise the way we live and identify ourselves.

The 27-year-old Bradford sisters are among more than 40 sets of identical twins around Britain who have been helping to trial the system known as Casper.

The technology analyses 32 different components of a telephone caller's voice - including gutteral sounds, intonation and how specific letters are pronounced - before producing a voice print.

It is more unique to that person than their fingerprint and can then be matched against information on other databases.

Mr Smith, managing director of the ICR Group, said while everyone's voice pattern was different, those of identical twins were the closest match of all, making them the ultimate test for the computer-based system's capabilities.

With Casper set to separate voices with a compatibility of up to 85 per cent it struggled to detect Lindsey using her twin's password but when the setting was raised to 88 per cent both voices were recognised immediately.

Fiona, who lives in Cottingley and is ten minutes older than her sister, confirmed the size of the challenge that faced Casper, saying: "Everyone gets confused - even mum and dad can't tell who it is when we just say 'hi it's me' and I don't think we could even tell each others' voices apart on the phone.

"I've heard Lindsey on answer machines and the only reason I know it's not me is because I know I didn't leave that particular message.''

Lindsey, of Nab Wood, said: "I've ended up chatting to Fiona's boyfriend for a few minutes before saying 'it's Lindsey' and he's said 'my God I thought it was Fiona'.

"I think this is a good idea - I knew we sounded alike but realised there must be slight differences and with technology as good as it is thought that it would find them and it has.''

Mr Smith said: "Most voices, even those of mimics, are easy to identify but for identical twins up to the age of about 18, living at home and with the same friends the differences are negligible and almost impossible to distinguish.

"But after that external influences such as having different friends and living or working in different places begin to have an affect. Subtle little differences creep in and though it's difficult for the human ear to pick them up Casper can.

"Lindsey and Fiona are in the hard to distinguish category but one went to college and the other didn't, while one works in Leeds and one in Bradford - there's a very slight difference in the inflection at the end of their sentences which Casper's picked up on.''

Mr Smith said they were now analysing the trial results to ascertain exactly which aspects of the voices were different, adding: "We haven't had any voices we've been unable to separate yet - the hardest were two 19-year-old identical twins and to get a guaranteed separation we had to raise the bar to 92 per cent.

"We're about to prove beyond reasonable doubt that even identical twins can be identified by their voices alone - that'll be a significant finding for speech technology and means there'll be a strong commercial viability for speech recognition solutions.''

Mr Smith predicts Casper will be in use commercially by the end of this year and envisages it will eventually be used to identify credit card holders, enabling totally secure banking and purchasing and drastically cutting credit card fraud.

Casper could also one day be used to identify telephone voters and those claiming benefits as well as replacing identity-related documents such as passports and driving licenses.