Machine that can turn our thoughts to words

A MIND-READING machine that turns thoughts into sentences will bring hope to millions of people left unable to communicate, such as stroke victims.

The device has an accuracy rate of 97 per cent – more than twice as high as other brain-signal decoding instruments.

Unlike existing machines which are limited to spelling words out slowly using residual eye movements or muscle twitches, the new device uses an algorithm to map the activity of neurons.

The brain signals are then fed into a computer which creates a representation of regularly occurring neural features.

Dr Joseph Makin, of California University in San Francisco, said: “These are likely to be related to repeated features of speech such as vowels, consonants or commands to parts of the mouth.

Decoded

“Another deep learning technique, known as a recurrent neural network, then decoded them into sentences. Average word error rates are as low as three per cent.”

Dr Makin and his team carried out their trials on four epilepsy patients who had been fitted with brain implants to monitor their seizures.

Their neural activity was turned “word by word into an English sentence – in real-time”, Dr Makin told the journal Nature Neuroscience.

In the last decade, brain-machine interfaces have enabled some amount of motor function to be restored to paralysed patients.

However, while mind-reading technologies are primarily designed to help the sick, ethical concerns have been raised.

Some experts say they have the potential to be misused and could be used to track people’s thoughts on behalf of governments or private companies.