Research
Meta's Brain2Qwerty v2 Advances Non-Invasive Brain-Reading
The Rundown AIhigh confidence
Why it matters
- →Non-invasive brain-computer interfaces could become more accessible, reducing the need for surgical implants.
- →Open-sourcing the technology allows other researchers to contribute and innovate further.
- →Improved accuracy in decoding brain signals enhances potential applications for individuals with speech impairments.
Meta has unveiled version 2 of its Brain2Qwerty system, which can decode full sentences from non-invasive brain scans with a 61% accuracy rate. This is a significant improvement over previous non-invasive methods and approaches the accuracy of surgical brain-computer interfaces. The system was tested with nine volunteers, generating nearly 22,000 sentences of data. Meta's decision to open-source the code and dataset could spur further research and development in non-invasive communication technologies.
Read originalThe story around this
Earlier coverage that leads up to this article, and what followed. Lines connect each piece to the closest one after it, converging here.
