Overview

• The sEMG-RD wristband employs surface electromyography sensors and neural networks trained on data from thousands of users to decode muscle signals without individual calibration.
• Prototype functions include real-time cursor movement, pinch and swipe gesture commands, and handwriting transcription at about 20.9 words per minute.
• Meta demonstrated the device controlling its Orion AR glasses for hands-free messaging, app navigation, and cursor control without a keyboard or mouse.
• Researchers at Carnegie Mellon are conducting trials with spinal-cord injury patients to verify that faint muscle activity can enable device control despite limited or no arm movement.
• Reality Labs has published a public repository of more than 100 hours of sEMG recordings from 300 participants to accelerate broader research into neuromotor interfaces.