Neural Prosthetics for Speech Restoration

In collaboration with Dr. Philip Kennedy and colleagues at Neural Signals, Inc., we created the world’s first brain-computer interface for real-time speech generation in 2007. This system decoded neural signals recorded from an electrode implanted in the speech motor cortex of a human volunteer with locked-in syndrome (characterized by full consciousness with near-complete paralysis) while he imagined speaking. This work received widespread coverage in the popular media, including pieces by CNN, the Boston Globe, and Esquire Magazine. A schematic of the brain-computer interface is provided below; further details about the system can be found in Guenther et al. (2009).


Schematic of a brain-computer interface for real-time speech generation (Guenther et al., 2009).

In more recent work, we developed a novel speech synthesizer that provides natural-sounding speech using a simple 2-dimensional input that can be generated via brain signals or muscle signals collected with electromyography (EMG). The video below, created by members of the Stepp Lab at Boston University, shows a user controlling this synthesizer via EMG signals from the face.