# Accurate Decoding of Natural Sentences from Non-Invasive Brain Recordings

arXiv: https://arxiv.org/abs/2608.18114 · HTML: https://mlsanity.com/paper/2608.18114/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 58 references that are in the map.
- First posted on arXiv 29 Jun 2026.

## Builds on (most cited first, 21 of 21)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Sigmoid Loss for Language Image Pre-Training](https://mlsanity.com/paper/2303.15343/) (2023)
- [Pointer Sentinel Mixture Models](https://mlsanity.com/paper/1609.07843/) (2016)
- [Optuna: A Next-generation Hyperparameter Optimization Framework](https://mlsanity.com/paper/1907.10902/) (2019)
- [Conformer: Convolution-augmented Transformer for Speech Recognition](https://mlsanity.com/paper/2005.08100/) (2020)
- [Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time](https://mlsanity.com/paper/2203.05482/) (2022)
- [SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition](https://mlsanity.com/paper/1904.08779/) (2019)
- [Decoding speech perception from non-invasive brain recordings](https://mlsanity.com/paper/2208.12266/) (2022)
- [Relaxing the Conditional Independence Assumption of CTC-based ASR by Conditioning on Intermediate Predictions](https://mlsanity.com/paper/2104.02724/) (2021)
- [The largest EEG-based BCI reproducibility study for open science: the MOABB benchmark](https://mlsanity.com/paper/2404.15319/) (2024)
- [Brain-to-Text Decoding: A Non-invasive Approach via Typing](https://mlsanity.com/paper/2502.17480/) (2025)
- [emg2qwerty: A Large Dataset with Baselines for Touch Typing using Surface Electromyography](https://mlsanity.com/paper/2410.20081/) (2024)
- [The 2025 PNPL Competition: Speech Detection and Phoneme Classification in the LibriBrain Dataset](https://mlsanity.com/paper/2506.10165/) (2025)
- [The Automated LLM Speedrunning Benchmark: Reproducing NanoGPT Improvements](https://mlsanity.com/paper/2506.22419/) (2025)
- [A foundation model of vision, audition, and language for in-silico neuroscience](https://mlsanity.com/paper/2605.04326/) (2026)
- [Time-Masked Transformers with Lightweight Test-Time Adaptation for Neural Speech Decoding](https://mlsanity.com/paper/2507.02800/) (2025)
- [NeuralSet: A High-Performing Python Package for Neuro-AI](https://mlsanity.com/paper/2605.03169/) (2026)
- [A cross-species neural foundation model for end-to-end speech decoding](https://mlsanity.com/paper/2511.21740/) (2025)
- [Assembling the Mind's Mosaic: Towards EEG Semantic Intent Decoding](https://mlsanity.com/paper/2601.20447/) (2026)

## Built on by (most cited first, 4 of 4)

- [The 2026 PNPL Competition: Word Classification and Efficient Cross-Subject Generalisation in LibriBrain100](https://mlsanity.com/paper/2609.03231/) (2026)
- [Myovox: Reading Speech from the Muscles of the Face](https://mlsanity.com/paper/2609.17548/) (2026)
- [Removing Timing Shortcuts Improves Non-Invasive Brain-to-Text](https://mlsanity.com/paper/2609.40359/) (2026)
- [Confidence-Ordering Reversal under Contextual Priors in Neural Decoding](https://mlsanity.com/paper/2610.08229/) (2026)

## Abstract (arXiv)

Restoring communication for people who have lost the ability to speak or move after a brain injury is a major challenge. While intracranial implants now enable high-performing brain-computer-interfaces, non-invasive alternatives are still lagging behind. Here, we present Brain2Qwerty v2, a model that can decode the production of natural sentences solely from real-time magnetoencephalography (MEG) recordings. By collecting 22,000 sentences typed by nine subjects, each recorded for 10 hours, our model leverages character, word and sentence-level representations to achieve an average word error rate (WER) of 39%. For our best participant, the model accurately decodes half of the sentences with one word error or less. Critically, decoding accuracy log-linearly improves with data volume, suggesting that the performance gap with intracranial approaches could be partially bridged through data scaling. We show that AI enables this performance in three main ways: the substitution of hand-crafted pipelines for event detection with deep learning, the finetuning of large language models to extract semantic representations, and the deployment of AI agents to iteratively refine our decoding pipeline via automated code development. Together, these results show that non-invasive brain-to-text decoding starts to operate at a level of accuracy previously thought exclusive to surgical implants, opening a path toward safe and efficient brain-computer-interfaces.