An NIH-funded research team demonstrated a brain-computer interface that can decode both speech and upper-body gestures from people with paralysis at the same time.
Researchers at the University of California, San Francisco used machine-learning systems to translate patterns of neural activity into spoken output and movement of a personalized digital avatar. Earlier interfaces typically focused on speech or movement separately.
The work remains experimental, but it points toward communication systems that preserve more of the nonverbal expression people naturally use during conversation.