In a stunning breakthrough, a severely paralyzed woman has been able to regain her ability to speak after 18 years, thanks to the power of artificial intelligence (AI). By translating her brain signals into speech and facial expressions, this groundbreaking technology has opened up a world of communication for individuals who were previously unable to express themselves naturally.
The patient, Ann, has been paralyzed since suffering a brainstem stroke over 18 years ago. In the past, she relied on slow speech synthesizers that required her to spell out words using eye tracking or small facial movements. However, these methods were time-consuming and made natural conversation nearly impossible.
With the help of a paper-thin rectangle implanted with tiny electrodes on the surface of her brain, Ann’s electrical brain activity is now detected and translated into speech and facial expressions by a digital avatar. This process allows her to communicate with others in a more embodied and natural manner.
The team at the University of California, San Francisco (UCSF), led by Professor Edward Chang, trained an AI algorithm to detect Ann’s unique brain signals for different speech sounds. By repeating various phrases, the computer was able to learn 39 sounds, which were then used to create a model that generates sentences based on the brain signals received. The avatar’s voice was also trained to sound like Ann, using recordings from before her injury.
Although the results are highly promising, there is still room for improvement. During testing, the system decoded words incorrectly around 28% of the time out of 500 phrases checked. However, the system’s current rate of generating brain-to-text is an impressive 78 words per minute, compared to the average rate of spoken conversation, which ranges from 110 to 150 words per minute.
This revolutionary technology not only gives hope to individuals like Ann but also opens up opportunities for people with severe paralysis to communicate more effectively in various professional fields. With further advancements, the potential for a full, embodied way of communication is within reach.
1. How does this technology work?
The technology works by implanting a paper-thin rectangle of electrodes onto the surface of the patient’s brain. These electrodes detect the electrical activity in the brain region responsible for speech and facial movements. The signals are then translated into speech and facial expressions by a digital avatar.
2. How accurate are the results of this technology?
During testing, the system decoded words incorrectly around 28% of the time out of 500 phrases checked. However, the system’s current rate of generating brain-to-text is 78 words per minute, which is still impressive considering the natural conversation rate of 110 to 150 words per minute.
3. What are the potential applications of this technology?
This technology has the potential to revolutionize communication for individuals with severe paralysis. It opens up opportunities for more natural and efficient communication in various professional fields, allowing paralyzed individuals to pursue careers as counselors, educators, and more.