Meta AI's Brain2Qwerty v2 decodes brain activity to text with 61% accuracy

Meta AI has developed Brain2Qwerty v2, an advanced system that decodes brain activity into text using non-invasive recordings. This new version achieves a 61% word accuracy rate, a significant improvement over previous non-invasive methods, and approaches the performance of surgical techniques. The research team is releasing the training code for both Brain2Qwerty v1 and v2, along with a dataset from their partner, the Basque Center on Cognition, Brain, and Language (BCBL), to foster open neuroscience research and aid individuals with communication impairments. AI

IMPACT Advances non-invasive brain-computer interfaces, potentially restoring communication for millions with neurological disorders.

RANK_REASON Research paper detailing a new AI model for decoding brain activity into text. [lever_c_demoted from research: ic=1 ai=1.0]

Read on Hacker News — AI stories ≥50 points →

• Basque Center on Cognition, Brain, and Language (BCBL)
• Brain2Qwerty v1
• Brain2Qwerty v2
• Digital Brain Project
• Electrocorticography
• Meta AI
• NeuralBench
• NeuralSet
• Stereotactic electroencephalography
• TRIBEv2

• product
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →