Code for "MEG-XL: Data-Efficient Brain-to-Text via Long-Context Pre-Training" at ICML 2026
A brain-computer interface (BCI) for decoding speech phonemes from EEG signals using a hybrid CNN-LSTM model, with interactive Streamlit visualizations and Docker support.
This repostiry contains the code requires to train a SVM and LDA models to distinguish between perceived speech and silence from fNIRS signals.
Syllable-balanced Mandarin reading corpus for BCI speech decoding
Consumer brain-computer interface for inner speech decoding. 8-channel EEG headband ($800) outperforms 128-channel clinical systems ($50K). EEGNet 35.5% accuracy (p=0.0006), cross-subject generalization (p=0.003). Real-time demo included.
How much a speech brain-computer interface decoder degrades after being trained once on early sessions, what changes in the neural signal, and whether small labeled recalibration can restore it. 20 months, 45 sessions, T15 Brain-to-Text '25 dataset.
Independent brain-to-text reproduction and research: speechBCI baselines, causal decoding, cross-day adaptation, and evidence-led experiments.
To associate your repository with the speech-decoding topic, visit your repo's landing page and select "manage topics."