Particle.news

Download on the App Store

Brain Implant Enables Paralyzed Woman to Speak in Real Time

New AI-powered technology translates neural signals into speech within one second, offering natural communication for individuals with severe speech paralysis.

Researchers connect Ann's brain implant to the voice synthesizer computer. (Photo by Noah Berger)
Representational image: This technology offers significant potential to improve the lives of individuals with severe speech paralysis.
digital brain image

Overview

  • Researchers tested the brain implant on a 47-year-old woman, Ann, who lost her ability to speak after a brainstem stroke 18 years ago.
  • The system uses AI to decode neural signals from the motor cortex and synthesize speech in near real-time, achieving unprecedented fluency.
  • Speech output is personalized to mimic the participant's pre-injury voice, trained on recordings from her past, such as her wedding video.
  • The technology eliminates delays seen in earlier systems, enabling fluid, natural conversations with a vocabulary of over 1,000 words.
  • While the system remains experimental, researchers are focusing on refining accuracy, expanding vocabulary, and incorporating emotional expressiveness.