A paralysed man has 鈥渟poken鈥� three different vowel sounds using a voice synthesiser controlled by an implant deep in his brain.
If more sounds can be added to the repertoire of brain signals the implant can translate, such systems could revolutionise communication for people who are completely paralysed.
鈥淲e鈥檙e very optimistic that the next patient will be able to say words,鈥� says , a neuroscientist at Boston University who led the study along with Philip Kennedy at s, a firm based in Duluth, Georgia, that produces neural implants.
Advertisement
Conventional speech
Eric Ramsey is 26 and has locked-in syndrome, in which people are unable to move a muscle but are fully conscious.
A brain implant, which requires invasive surgery, may sound drastic. But lifting signals directly from neurons may be the only way that locked-in people like Ramsey, or those with advanced forms of , a neurodegenerative disease, will ever be able to communicate quickly and naturally, says Guenther.
Devices that rely on interpreting residual muscle activity, such as eye blinks, are no good for people who are completely paralysed, while those that use brain signals captured by scalp electrodes are slow, allowing typing on a keyboard at a rate of one to two words per minute.
鈥淥ur approach has the potential for providing something along the lines of conventional speech as opposed to very slow typing,鈥� he says.
Messy signals
His team鈥檚 breakthrough was to translate seemingly chaotic firing patterns of neurons into the acoustic 鈥渂uilding blocks鈥� that distinguish different vowel sounds. Ramsey, who suffered a brain-stem stroke at the age of 16, has an electrode implanted into a brain area that plans the movements of the vocal cords and tongue that underlie speech.
Over the past two decades, the team has developed models that predict how neurons in this region fire during speech. Using these predictions, they were able to translate the firing patterns of several dozen brain cells in Ramsey鈥檚 brain into the acoustical building blocks of speech.
鈥淚t鈥檚 a very subtle code; you鈥檙e looking over many neurons. You don鈥檛 have one neuron that represents 鈥榓aa鈥� and another that represents 鈥榚ee鈥�. It鈥檚 way messier than that,鈥� Guenther says.
Next, Guenther鈥檚 team provided Ramsey with audio feedback of the computer鈥檚 interpretation of his neurons, allowing him to tune his thoughts to hit a specific vowel. Over 25 trials across many months, Ramsey improved from hitting 45聽per cent of vowels to 70聽per cent.
Listen: Eric Ramsey repeating target vowel sounds here
Laptop control
Of course, the ability to produce three distinct vowels from brain signals won鈥檛 allow for much communication, let alone real-time natural conversation. But Guenther says technological improvements should have a next-generation decoder producing whole words in three to five years.
This next device will read from far more neurons and so should be able to extract the brain signals underlying consonants, says Guenther. The team plan to have it controlled by a laptop, so people can practise speaking at home as much as they like. Two people interested in having this device implanted have already contacted Guenther鈥檚 team, he says.
, a neuroscientist at the University of T眉bingen, Germany, who has developed a prosthetic that records brain activity beneath the skin to type out words, is sceptical that the new approach will yield fluent speech.
He also worries about its reliance on an invasive brain surgery. 鈥淚n most cases an invasive procedure like this where you hurt the brain is not necessary,鈥� he says.
Guenther agrees that 鈥渋f patients have enough residual movement they can control some sort of device鈥�. But he says that his implant is intended principally for people who suffer from severe forms of paralysis or significant vocal tract damage that means even these interventions won鈥檛 work.
Journal reference:
