JAN 31, 2019 12:57 PM PST

Translating Brain Signals Directly into Speech

WRITTEN BY: Nouran Amin

In Scientific Reports, neuro-engineers developed a system that translates thought into intelligible and recognizable speech through detections of brain activity that ultimately reconstruct the words a person hears with unprecedented clarity. The system can help people who cannot speak, those living with amyotrophic lateral sclerosis (ALS), or individuals recovering from stroke.

Concept illustration (stock image) via Science Daily  Credit: © adragan / Fotolia

"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," says Nima Mesgarani, PhD, the paper's senior author and a principal investigator at Columbia University's Mortimer B. Zuckerman Mind Brain Behavior Institute. "With today's study, we have a potential way to restore that power. We've shown that, with the right technology, these people's thoughts could be decoded and understood by any listener."

Previous research has confirmed that when someone speaks or even thinking of saying something—patterns of brain activity begin to surface. Early efforts of the study focused on decoding brain signals through the utilization of computer models that analyzed spectrograms (the visual representations of sound frequencies). However, this approach has failed and so the study shifted to the use of a ‘vocoder’ which is a computer algorithm capable of synthesizing speech after being trained on voice recordings of human conversations

"This is the same technology used by Amazon Echo and Apple Siri to give verbal responses to our questions," said Dr. Mesgarani, who is also an associate professor of electrical engineering at Columbia's Fu Foundation School of Engineering and Applied Science. "Working with Dr. Ashesh Dinesh Mehta (co-author of the study), we asked epilepsy patients already undergoing brain surgery to listen to sentences spoken by different people, while we measured patterns of brain activity," said Dr. Mesgarani. "These neural patterns trained the vocoder."

Source: The Zuckerman Institute at Columbia University

About the Author
  • Nouran earned her BS and MS in Biology at IUPUI and currently shares her love of science by teaching. She enjoys writing on various topics as well including science & medicine, global health, and conservation biology. She hopes through her writing she can make science more engaging and communicable to the general public.
You May Also Like
APR 23, 2020
Chemistry & Physics
Wearable e-skin powered by sweat
APR 23, 2020
Wearable e-skin powered by sweat
What if your skin could monitor your heart rate, body temperature, and blood sugar levels? Well, not your skin, exactly ...
MAY 11, 2020
Space & Astronomy
Hear What Astronauts Think About SpaceX's Upcoming Crewed Launch
MAY 11, 2020
Hear What Astronauts Think About SpaceX's Upcoming Crewed Launch
On Wednesday, May 27th, NASA will entrust commercial space company SpaceX with the coveted task of flying astronauts to ...
MAY 15, 2020
Technology
What is HARVEY?
MAY 15, 2020
What is HARVEY?
One of the biggest challenges facing clinical workers is trying to explore user interface treatment options easily witho ...
MAY 28, 2020
Chemistry & Physics
Expandable Resin Solves the Size Problem of 3D Printing
MAY 28, 2020
Expandable Resin Solves the Size Problem of 3D Printing
One of the current limitations of using a 3D printer is the size of the product. For someone to print a large part, they ...
JUL 26, 2020
Drug Discovery & Development
AI Creates Proteins that Rival Those in Nature
JUL 26, 2020
AI Creates Proteins that Rival Those in Nature
Proteins are crucial for the function of cells in living organisms. Now, researchers have developed a machine leaning-le ...
JUL 22, 2020
Technology
Software Capabilities for Self-Driving Cars
JUL 22, 2020
Software Capabilities for Self-Driving Cars
Self-driving (autonomous) cars have already hit the road. The software capability of these cars, especially those that r ...
Loading Comments...