The brain style allows the person to speak and sing his ‘true voice’
A new brain-computer interface prints the thoughts of singing and expressing language in real time
The motor cortex (Orange, Illustration). Electrodes set in this region helped recording activity related to someone who cannot speak intelligent.
KATERYNA KONEY / SCIENCE PHOTO LIBRARY / PHOTOS IN ALAMY STOCK
A person with a severe deficit of speaking is to speak openly and singing using a brain implant that translates his neural activity to the words nearby. The device gives changes to the tone when he asks questions, emphasizes the words he chose and allowed him to make a string to three pitches.
The system – known to be A brain interface (BCI) – Used artificial intelligence (AI) to minimize participant brain activity as he attempts to speak. The device is the first to change not only to the intended words of a person but also have natural language such as tone, pitch and weight to express meaning and emotion.
In a study, a synthetic voice imitated by the participant’s owner in his words within 10 milliseconds of neural activity signing his purpose in speaking. The system, now described in NATUREmarks an important development of the first BCI models, streaming speech for three seconds or it is made after users finished wearing a whole sentence.
In support of science journalism
If you enjoy this article, think about supporting our winning journalism in Subscribe. By purchasing a subscription you helped to ensure the future of influential stories about the discoveries and ideas that make our world today.
“This is the Holy Grail to speak BCIs,” Christian Herff, a computation neuroscientist at Masastricht University, Netherlands, not included in the study. “It is now true, strong, continuous speech.”
Real-time decoder
The participant of study, a 45-year-old man, lost his ability to speak clearly after developing Amyotrophic lateral scllersiss, a form of motor neuron saliwhich damage the nerves that control muscle movements, including the requirements for speech. Even if he can do the words and words in the mouth, his speech slowly and is unclear.
Five years after his symptoms began, the participant surgically surgery to insert 256 silicon electrodes, per 1.5-mm in length, in a region of the brain controlling activity. Studying the co-author Maitreyee Wairagie, a neuroscientist at the University of California, Davis, and his companions who repaired his brain signals every 10 milliseconds. Their system decodes, in real time, sounds seek to produce rather than his intended words or subunit phonemes – subunit words spoken words.
“We don’t always use words to indicate what we want. We have interviews. We have other expressionszoly exceptions to the vocabulary,” Wairagiwar explains. “To do that, we have adopted this procedure, completely unstoppable.”
The team also identifies the synthetic sound like oneself, by training AI algorithm in recordings of his illness interviews.
The team asked the participant to try to make interjections like ‘aah’, ‘ooh’ and ‘hmm’ and say the made words. BCI has successfully made these sounds, indicated that this language can be made without a fixed vocabulary.
Freedom of speech
Using the device, the participant wrote the meetings, responding to open questions and said what he wanted, using some words that were not about decoder training data. He told researchers who listened to the synthetic voice that made his talk that made him “feel happy” and it feels like his “real voice”.
In other experiments, BCI recognized whether the participant seeks to say a sentence as a question or as a statement. The system can also know if he stresses different words in the same sentence and adjust the tone of his synthetic voice. “We bring all the different elements of human speech that is very important,” said Wairagikar. Previous BCIS can only be flat, monotone speech.
“This is a little transfer to paradigm that it can lead to a real life tool,” says Silvia Marchestive, a Neuroengineer at the University of Geneva in Switzerland. System features “are important for adoption for daily use for future patients.”
This article has been copied with permission and first published On June 11, 2025.
