By Claire Bates
A revolutionary new device that reads a person’s thoughts and turns them into speech could soon change the lives of paralysed patients around the world.
The Neuralynx System is being developed by a team of scientists led by Professor Frank Guenther at Boston University.
Users will simply have to think of what they want to say and a voice synthesizer will translate the thoughts into speech almost immediately.
They have tested the device on a patient who has ‘locked-in syndrome’, after a stroke stopped neural signals travelling from his brain to the rest of his body.
The rare condition means the person is aware and awake, but cannot move or communicate due to complete paralysis of nearly all voluntary muscles in the body except for the eyes.
The 26-year-old volunteer was asked to think of a series of basic vowel sounds. The researchers were able to translate these and vocalise them in just a fraction of a second using the new system.
His accuracy increased with each practice session from 45 per cent to 89 per cent.
Scientists began the experiment three years ago, when they implanted an electrode in the patient’s brain on the boundary of the two regions that govern speech and movement.
Within four months, neurites had grown into the electrode and begun producing neural signals.
When the patient was asked to think of different vowel sounds the electrode acted to amplify and convert his thought signals into FM radio waves.
These were transmitted wirelessly to a receiver coil that had been attached to the patient’s head.
From here they were sent to a special recording system that turned them into digital data. this was then decoded and turned into speech by two programmes running on a computer.
The whole process took just 50 milliseconds, which meant it effectively worked in real time.
The system is far quicker than current methods used by those with severe physical disabilities such as Professor Stephen Hawking.
‘The results of our study show that a brain-machine interface user can control sound output directly, rather than having to use a relatively slow typing process,’ Professor Guenther said.
‘They also provide our first insight into how neurons in the brain represent speech.’
Professor Guenther said they next plan to develop a synthesizer that will be able to produce consonants as well as vowels.
He said: ‘We are working on hardware that will greatly increase the number of neurons that are recorded. We expect to tap into at least 10 times as many neurons in the next implant recipient, which should lead to a dramatic improvement in performance.’
Dr Christopher J James from Southampton University said: ‘The science is actually quite novel and although what has been produced is still in its infancy it clearly demonstrates the rehabilitative power of a Brain-Computer Interface.’