×
AI brain implant restores speech to locked-in stroke survivor after 18 years
Written by
Published on
Join our daily newsletter for breaking news, product launches and deals, research breakdowns, and other industry-leading AI coverage
Join Now

UC Berkeley and UC San Francisco researchers have successfully restored speech to Ann Johnson, a woman who lost the ability to speak after a brainstem stroke 18 years ago, using an AI-powered brain-computer interface. The breakthrough technology translates brain activity into speech in real-time, offering hope for people with locked-in syndrome and potentially transforming accessibility in the workforce and beyond.

What you should know: Johnson suffered a brainstem stroke at age 30 in 2005 that left her with locked-in syndrome—a rare condition causing near-complete paralysis and loss of speech while leaving cognitive abilities intact.

  • She joined the clinical trial in 2022 as the third participant in a study that began in 2020.
  • The technology uses a neural implant that reads brain activity from the speech processing region and translates it into audio and visual communication through an AI model.
  • Johnson can currently communicate at about 14 words per minute using eye-tracking technology, compared to normal conversational speech of 160 words per minute.

How it works: The neuroprosthesis reads brain signals when someone attempts to speak and translates them into audio and avatar movements.

  • An implant rests on the brain region that processes speech, acting as a “thought decoder” that captures neural signals when Johnson tries to speak.
  • The AI model translates these brain signals into text, audio, or facial animations, similar to how Siri converts voice to text.
  • Researchers used a recording of Johnson’s wedding speech to recreate her voice and allowed her to choose from different avatars for visual representation.
  • The system only activates when someone makes a deliberate effort to speak—it cannot read random thoughts.

Major breakthrough: The team dramatically reduced response delays from eight seconds to just one second using new streaming architecture.

  • The original 2023 system required users to attempt entire sentences before translation, using sequence-to-sequence architecture.
  • The March 2024 update introduced streaming architecture that translates brain activity to sound in real-time with only about a one-second delay.
  • This advancement was published in Nature Neuroscience and represents a significant step toward natural conversation flow.

What they’re saying: The emotional impact of hearing her own voice again was profound for Johnson and the research team.

  • “What do you think of my artificial voice?” Johnson asked during the trial. “Tell me about yourself. I am doing well today.”
  • “We didn’t want to read her mind,” said Gopala Anumanchipalli, UC Berkeley assistant professor of electrical engineering and computer sciences. “We really wanted to give her the agency to do this.”
  • Johnson wrote to researchers: “I want patients there to see me and to know their lives are not over now. I want to show them that disabilities don’t need to stop us or slow us down.”

Future vision: Researchers envision plug-and-play neuroprostheses becoming standard medical care within years.

  • “We need to be able to have neuroprostheses be plug-and-play, so that it becomes a standard of care and not a research experiment,” Anumanchipalli explained.
  • Future developments could include wireless implants, 3D photorealistic avatars, and digital clones with personalized preferences.
  • Johnson hopes to become a counselor in physical rehabilitation facilities, using the technology to communicate with clients.

The bigger picture: While the target population is relatively small, researchers emphasize these patients are among the most vulnerable in terms of quality of life.

  • The technology has “enormous potential to make the workforce and the world more accessible to people like Johnson,” according to the research team.
  • Johnson had her implant removed in February 2024 for unrelated reasons but continues collaborating with researchers on future improvements.
  • The breakthrough represents a significant advance in restoring human communication abilities through direct brain-computer interfaces.
A stroke stole her ability to speak. Eighteen years later, scientists used AI to bring it back.

Recent News

Amazon slashes AI voice recorder price 46% to all-time low

The MagSafe-compatible device combines ChatGPT 5.0 with transcription across 100+ languages.

Anthropic to pay $1.5B in first US AI copyright settlement

The company paid roughly $3,000 per pirated book from shadow libraries like LibGen.

Lenovo launches AI-powered Android tablets starting at $270

Lenovo bets AI belongs in budget tablets, not just premium ones.