
Meta's Brain2Qwerty decodes speech from brain signals without implants
Hacker News·4mo·alok-g
Researchers at Meta trained an AI model to translate non-invasive brain signals directly into text, potentially opening a communication pathway for paralyzed or speech-impaired users without requiring surgical implants. The approach uses standard EEG equipment, which could make the technology more accessible than previous brain-computer interface work that relied on invasive electrode arrays.
Original story
Read the original on Hacker NewsRelated stories


Devtools
Espressif releases ESP32-S31, a stripped-down microcontroller for cost-conscious projectsHacker News·5mo·volemo