Research

Meta Unveils Brain2Qwerty v2 for Non-Invasive Brain-to-Text Communication

Meta has released Brain2Qwerty v2, an AI system achieving 61% word accuracy in decoding non-invasive brain signals into text, surpassing previous non-surgical methods.

High-tech laboratory equipment with computer system in lab setting.

Photo: Media Dung / Pexels

Meta has announced the release of Brain2Qwerty v2, an advanced AI system designed to decode brain activity into text without the need for surgical implants. This latest version represents a significant leap forward in non-invasive brain-to-text communication, with the system now capable of real-time sentence decoding from raw brain signals. The breakthrough comes after a year of research and development, building on the initial release of Brain2Qwerty v1. The new model is the highest-performing end-to-end pipeline for this task, approaching accuracy levels previously reserved for invasive techniques.

The system was trained on approximately 22,000 sentences from nine volunteer participants, each recorded for 10 hours while wearing a magnetoencephalography (MEG) device and actively typing. Unlike earlier methods that relied on hand-crafted pipelines, Brain2Qwerty v2 uses end-to-end deep learning to directly decode raw brain signals. Fine-tuning large language models on neural data enables the system to leverage semantic context, bridging the gap between noisy brain recordings and coherent language. AI agents were also deployed to explore optimizations for the decoding pipeline, with final configurations selected manually by engineers.

The research is part of Meta's broader efforts to develop open foundational models of the brain, in collaboration with the Basque Center on Cognition, Brain, and Language (BCBL). The company has released the full training code for both v1 and v2, as well as the v1 dataset, to accelerate neuroscience breakthroughs. The goal is to improve communication for individuals with brain lesions and to help bridge the gap between non-invasive and invasive neuroprosthetic approaches.

Source: metaai

Key points

Meta has released Brain2Qwerty v2, the highest-performing end-to-end pipeline for non-invasive brain-to-text communication.
Brain2Qwerty v2 achieves a word accuracy rate of 61%, significantly improving upon the 8% word accuracy from other non-invasive methods.
The system was trained on approximately 22,000 sentences from nine volunteer participants, each recorded for 10 hours wearing a magnetoencephalography (MEG) device while actively typing.
Fine-tuning large language models on neural data allows the system to leverage semantic context, bridging the gap between noisy brain recordings and coherent language.
Decoding accuracy improves log-linearly with data volume, suggesting the remaining performance gap with surgical approaches could be narrowed through data scaling alone.

Source: Meta AI Read the original →

WRITTEN BY

Maya Chen

AI Research & Breakthroughs

Maya breaks down the latest AI research papers, benchmarks, and technical breakthroughs into plain language.