Meta Wristband Interface: AI Translates Brain Signals

Imagine the ability to control machines with your mind, rather than having to write on a keyboard or click on the mouse. Now the parent company on Facebook aims to the best thing next – a new wrist that, with the help of artificial intelligence, is the conclusion of electrical commands sent from the brain to the muscles and converting them into computer signals, all in an expanded way. Although experts doubt that it will replace keyboards and mice for traditional computing, it may have new uses for a wide range of applications, such as wearable interfaces for mobile devices, or auxiliary technologies that are controlled by thinking for persons with disabilities.

The bracelet is used from the Meta Labs in Meta Metal Communications placed on the skin to detect electrical signals from the muscles – a technique known as the surface -to -surface layout (SEMG) – which is created in response to orders from the brain. The very sensitive system transmits this data to a computer using Bluetooth to help it identify gestures such as signal and disk in actual time, and detailed results in nature on July 23.

The bracelet is not a direct interface with the brain. “It is not a system of mind reading. It cannot make you behave differently as your will imposes, it does not link you to other nervous people, and does not predict your intentions,” says Dario Farina, the president of Neurororebility at the Empire College of London, who has not participated in dead research but tested technology. (Meta was unable to make anyone to comment until the time of the press.)

How can Amnesty International a dead wrist

The previous “Neuromotor” devices, such as the suspended Myo Myo, sought to use the SEMG for the computer. One of the main challenges faced by these previous devices is how each of them needs a long -time personal calibration for every user to calculate the differences in human anatomy and behavior.

https://www.youtube.com/watch? See how the device detects thumb, fingers, and handwriting gestures.Labs of reality in Meta

In contrast, Mita says her bracelet can work mainly. The key was to train artificial intelligence systems on deep learning on data from more than 6000 paid volunteers wearing the device. Joe Paradeso, head of the MIT Media Laboratory Research Group, who has not participated in this study, says these models created and that can accurately explain the user’s inputs through different people without the need for individual calibration, says Joe Paradeso, head of the MIT Media Laboratory, who has not participated in this study.

“The amount of information that has been decoded with this device is very large, much larger than any previous attempt,” says Varina. “The device can recognize the handwriting, for example, which could not have been imagined before.”

The surface electrical imaging is used to measure the non -gaseous electrical activity of the muscles that control manual gestures.Labs of reality in Meta

There is a possible source of anxiety to use this wrist is how users may not want to explain every manual movement as inputs – if they have to scratch itching, or pick up a cup of water. However, the device is trained to identify certain orders only, while ignoring other gestures. “The potential user of the device can perform any daily life activity without risk to activate the device, however controlling the device with the specific gestures for which the device has been trained,” says Varina.

In tests, volunteers who have never tried the bracelet can use hand writing gestures to enter a text at 20.9 words per minute. (Compared, the average speed of writing the mobile keyboard is about 36 words per minute.)

“I doubt that most computer users will rush to buy this wrist if it becomes commercially available,” says Ibn Kirksy, a professor of anthropology at Oxford University who studies the interaction between science, technology and society. “Most people write about 40 words per minute, and very skilled printing can exceed 100 words per minute. Since this new wearable device allows users only to write 20 words per minute, and I doubt that many people want to spend enough time to learn how to use this new method of interface with computers.”

Instead, the new study argues that the bracelet may be useful in scenarios where keyboards or mice are limited, such as portable and wearable applications. “It is not a alternative to the keyboard,” Paradeso says.

For example, when it comes to virtual reality and the enhanced reality glasses followed by Meta, Google and others, the preachers of the previously photographed these devices include the followers based on vision who follow the hands of the hands that were placed in front of users. However, the fatigue that can be launched with the “Gorilla Arm” approach limits long -term use, and “the need to use the space in front of you has its faults”, notes Paradiso.

Since this new bracelet explains the electrical signals instead of movement, it may allow users instead to keep their hands on their sides and interact with the devices using fine finger movements. Paradiso says: “Facades like these are needed for the glasses that arise everywhere,” Paradiso says. “My favorite scenario is in a crowded train with my worn outlets that was adhering to the news, or communicating with friends.

Researchers also suggest that the wrist can help people with disabilities better interact with computers, especially individuals who suffer from low movement, muscle weakness, fingers amputation, paralysis and more. “Some members of the disabled society face difficulty in writing on traditional keyboards, or the use of computer mice,” says Kirksy, who has not participated in this research. “This device offers a new option that may help some members of this society, who have very specific physical challenges, and face computers.”

For such applications, the virtue of the new bracelet is that it is relatively easy to not adhere and dof. On the other hand, the BCIs, which also depend on electrical signals from the brain, often require invasive brain surgery. Paradiso says: “There is no invasive BCIS, like those that apply electrodes to the scalp,” but its wrist is more comfortable than Skullcap, “Paradiso says.

However, many questions remain before this new technology may help people with disabilities. “This type of wearing is assumed to be a typical tip and precise control of the engine,” says Suleiman Romney, who was previously head of the Microsoft Comprehensive Design Laboratory. “As a person who is a horrific part of the limbs, I am always looking for ways to transfer activities to Limblet instead of continuing to increase the loading of my typical hand. How easy is it to suit non -stereotypes, muscles or muscles? How can the tremors be filtered?

Ultimately, “the main obstacle is widely published,” says Varina. “To be distributed to millions of individuals, the system must be strong through different anatomicals and to be used with the minimum error rate. To match the error rate of mouse or keyboard in the vast population, it is definitely a major challenge.”