Meta's Latest Wristwear Offers Direct Command Over Digital Devices
Researchers at Meta have made a groundbreaking advancement in digital interaction technology with the development of a wristband that reads brain signals sent to muscles, paving the way for a new era of low-effort control for computer interactions.
The wristband, which uses surface electromyography (sEMG) to detect electrical signals from motor neurons to muscle fibers, can interpret the user's intended finger and hand movements before muscles physically move. This innovative technology allows for control of digital devices like computers and augmented reality (AR) glasses without the need for traditional input devices such as keyboards, mice, or touchscreens.
The wristband is equipped with 48 gold-plated electrodes and employs deep learning models trained on data from thousands of volunteers to decode the intent of the user. The AI system within the wristband recognizes common electrical patterns across different users, enabling it to function without individual calibration and to work "out of the box" for new users.
The sEMG models show improvements as they train on more people, similar to how language models improve with more data. As more extensive training datasets are gathered and models are deployed in situations where individuals can customize sEMG models to their unique writing style, sEMG decoding models will continue to advance.
Meta is releasing a public dataset of over 100 hours of sEMG recordings from 300 participants for further development and refinement of the technology. With a small amount of personal data (just 20 minutes), the system's performance improved by 16%.
The technology has great potential for people with disabilities, particularly those with paralysis or other motor impairments. Because sEMG signals are stronger and faster than EEG-based brain signals, this technology can provide a more immediate, precise, and non-invasive interface for assistive communication and device control for disabled individuals.
Researchers are already looking ahead to accessibility applications, such as allowing someone with limited mobility to control a phone, computer, or robotic tools using only tiny, effortless muscle signals. The typing speed with the device is currently 20.9 words per minute, slower than the average smartphone typing speed of over 33 words per minute, but improvements are expected as the technology continues to evolve.
In collaboration with Carnegie Mellon University, Meta has tested the wristband with users who have spinal cord injuries and hand paralysis, where some residual muscle activity remains but physical movement is limited or absent. Personalizing handwriting models with minimal fine-tuning for individual participants can lead to a 30% improvement in performance.
The system works for all users and does not need to be trained for each individual. Users can control the computer by making gestures such as swipes, pinches, and taps, and can even write letters in the air. The device, called surface electromyography (sEMG), detects electrical signals from muscles and communicates with a computer via Bluetooth.
This advanced, non-invasive approach could significantly improve accessibility and digital interaction for people with motor impairments, making computing and communication more intuitive and effortless.
- This groundbreaking advancement in digital interaction technology, developed by researchers at Meta, utilizes a wristband equipped with 48 gold-plated electrodes that reads brain signals sent to muscles.
- The technological innovation allows for the control of digital devices, such as computers and augmented reality glasses, without the need for traditional input devices like keyboards or touchscreens.
- The wristband's AI system, employing deep learning models, recognizes common electrical patterns across different users, eliminating the need for individual calibration.
- As more extensive training datasets are gathered and models are deployed, the sEMG decoding models will continue to advance, potentially revolutionizing medical-conditions management for those with paralysis or other motor impairments.
- The technology, dubbed surface electromyography (sEMG), will not only provide a more immediate, precise, and non-invasive interface for assistive communication and device control for disabled individuals but could significantly improve accessibility and digital interaction for people with motor impairments.
