A team of researchers has been working on understanding how our brains control speech and using that knowledge to create technology that can help people who have lost their ability to speak.
They used complex computer programs to study brain activity and recreate speech from brain recordings.
This technology can analyze the brain processes that drive human speech.
How it works
Speech is a complicated process involving different parts of our brain.
Researchers wanted to figure out how our brains control our vocal muscles and process the sound of our own voices when we speak.
They used a deep learning method to decode speech parameters from brain signals.
The researchers found that our brain processes feedback (hearing our own voice) and feedforward (telling our muscles how to move) signals at the same time during speech.
This challenges the idea that these processes are completely separate in the brain.
Using this new understanding, the researchers are developing prostheses (devices) that can read brain activity and turn it into speech.
What's unique about their approach is that it can recreate the patient's own voice, and it only needs a small amount of data to do so.
They collected data from epilepsy patients with electrodes in their brains to better understand how speech is controlled in the brain.
Why it's important
This research could potentially help people who have lost their ability to speak by allowing them to regain their own voices through technology that deciphers their brain signals and turns them into spoken words.