- Meta launches Brain2Qwerty v2 AI system.
- It will convert brain’s thoughts into text without surgery.
- It will be useful for paralyzed and unable to speak people.
- The word accuracy of the system is up to 61%, quite high.
Meta Brain2Qwerty v2: Meta has created an AI system that will be able to read the thoughts going on in your mind. This system has been named Brain2Qwerty v2 and it will convert brain activity into text. That is, this system can present the thoughts going on in your mind in text format on the screen even without you telling. Its special thing is that neither surgery is required nor will it have to be implanted. This AI system will be very useful for people who are paralyzed or unable to speak due to any other condition. This is the second version of Meta’s Brain2Qwerty AI system introduced last year.
How will this system work?
Meta said that this system uses magnetoencephalography (MEG). Such systems do not require surgery. With the help of a helmet fitted with sensors, it measures the magnetic fields produced by neural activities in the brain. If it is understood in simple language, this system works with the help of a special helmet that looks like a hair dryer. There are many sensors installed in this helmet, which catches signals of brain activity. It is special because it does not require surgery or implants to use it. For this reason its use is safe and cheap.
Many more features in the new version
According to Meta, Brain2Qwerty v2 has been trained with more than 22,000 sentences from 9 volunteers. Each of these volunteers wore this system for 10 hours and typed the thoughts going on in their mind. Compared to the old version, the new version uses end-to-end deep learning to decode language only from the signals going on in the brain. This system itself can recognize the pattern of brain signals with the help of AI.
How accurate is the new system?
Talking about accuracy, Meta claims that its per word accuracy is 61 percent and among the best performing participants it had reached 78 percent. The accuracy of other systems of this method is only up to 8 percent. To make the system better and efficient, Meta has also fine-tuned the Large Language Model so that it can capture the grammatical context of the conversation.
Read this also-
Now buying iPhone has also become expensive, the price rise has started here
