Related Experiment Video
Updated: Jun 28, 2025
06:04
Systematic Hearing Performance Evaluation Process for Adolescents with Cochlear Implantation at Early Ages
Published on: March 24, 2023
382
An accurate and rapidly calibrating speech neuroprosthesis
Nicholas S Card1, Maitreyee Wairagkar1, Carrina Iacobacci1
1Departments of Neurological Surgery, University of California Davis, Davis, CA, USA.
Medrxiv : the Preprint Server for Health Sciences
|April 22, 2024
Summary
A novel speech neuroprosthesis enables rapid communication for individuals with paralysis by decoding attempted speech from brain activity. This brain-computer interface achieved high accuracy with minimal training, restoring naturalistic communication for an ALS patient.
Area of Science:
- Neuroscience
- Biomedical Engineering
- Assistive Technology
Background:
- Brain-computer interfaces (BCIs) offer communication potential for individuals with paralysis.
- Existing BCIs face challenges with extensive training data and word accuracy.
Approach:
- A speech neuroprosthesis was implanted in a participant with ALS and severe dysarthria.
- Intracortical electrodes recorded neural activity associated with attempted speech.
- A system decoded neural signals into text and synthesized speech.
Key Points:
- High accuracy (99.6%) achieved with a 50-word vocabulary after 30 minutes of training.
- 90.2% accuracy reached with a 125,000-word vocabulary after 1.4 additional hours of training.
- Sustained 97.5% accuracy beyond eight months, enabling over 248 hours of self-paced conversation.
Conclusions:
- An intracortical speech neuroprosthesis can restore naturalistic communication.
- Minimal training data is required for high performance.
- This technology significantly improves communication for individuals with severe speech impairments.
