Artificial intelligence (AI) is becoming more powerful and enjoys greater applications in various scientific disciplines. Now researchers from Columbia University came up an idea for an AI-powered gadget which can translate thoughts into speech. Not only is this a great invention, but it can help people who are unable to speak.
“Our voices help connect us to our friends, family, and the world around us, which is why losing the power of one’s voice due to injury or disease is so devastating,” researcher Nima Mesgarani said in a statement.
He added that application of the right technology may be the way to “restore that power.” Using technology, scientists could decode thoughts and translate them into understandable speech. Their paper published in the journal Scientific Reports describes how the team used an AI-powered gadget capable of producing sounds which greatly resemble human speech.
“This is the same technology used by Amazon Echo and Apple Siri to give verbal responses to our questions,” Mesgarani said.
To build the AI-powered gadget which can translate thoughts into speech, the team asked five people suffering from epilepsy who were scheduled to undergo brain surgery to listen to sentences while researchers looked at a neural network of their brain activity. They then used those patterns to teach their AI-powered algorithm to read them, calling it “vocoder.” After learning the patterns, the algorithm was able to translate brain waves into speech.
The researchers then asked the patients to listen to someone count from zero to nine while their brain signals were being recorded and then fed those recordings into the algorithm. Finally, the team found 11 people to listen to the speech to test the device’s effectiveness.
“We found that people could understand and repeat the sounds about 75% of the time, which is well above and beyond any previous attempts,” Mesgarani said. “The sensitive vocoder and powerful neural networks represented the sounds the patients had originally listened to with surprising accuracy.”
Currently the AI-powered gadget can only translate thoughts while someone listens to speech. However, the success of the study so far has encouraged the team to continue working on the study and improving their algorithm. In the future, they expect people to be able to speak or think about speaking without having to listen to anything.
“In this scenario, if the wearer thinks, ‘I need a glass of water,’ our system could take the brain signals generated by that thought, and turn them into synthesized, verbal speech,” Mesgarani said.
Mesgarani added that the enhancement of the algorithm which can translate thoughts into speech will change the lives of many people who are unable to speak for various reasons. People who suffer from certain diseases or injuries would have a much easier time communicating with those in their surroundings, and they would be able to connect with the rest of the world. The invention has major applications in medicine, but it could certainly be applied to other scientific disciplines as well.