They create the first mind-reading helmet capable of translating brain waves into words

Australian scientists have developed what would be the world’s first artificial intelligence capable of reading minds and translating brain waves into readable text.

To do this, they use a helmet covered in sensors, which are capable of observing the electrical activity in the brain.

Researchers at the University of Technology in Sydney invited several participants to read text while wearing this particular helmet packed with sensors, which was dedicated to recording the electrical activity of the brain.

These EEG recordings were then converted to text using an artificial intelligence model, called DeWave.

Although the system is still not perfect, in fact It has an accuracy of only 40%the researchers comment that their intention is for this percentage to grow almost to perfection.

The DeWave model was trained by observing examples where brain signals match specific sentences.

“For example, when you think about saying ‘hello,’ your brain sends certain signals,” he says. Charles Zhou of the UTS. “DeWave learns how these signals relate to the word ‘hello’ by seeing many examples of these signals for different words or sentences.”

Once DeWave had a good understanding of the brain’s signals, the researchers plugged it into an open-source LLM language model, similar to the AI ​​that powers ChatGPT.

“This LLM is like a smart writer who can form sentences. We tell this writer to pay attention to DeWave’s signals and use them as a guide to create sentences,” Zhou says.

“It is the first to incorporate discrete coding techniques into the brain-to-text translation process, introducing an innovative approach to neural decoding. Integration with large linguistic models is also opening new frontiers in neuroscience and AI,” he adds. Chin-Teng Linfrom the University of Technology Sydney (UTS).