Transform Brain Signals
Into Natural Language
NEST is an open-source deep learning framework that decodes EEG brain activity into readable text with state-of-the-art accuracy.
From Brain Activity to Text
NEST uses a transformer-based architecture to decode EEG signals in real-time
EEG Recording
Capture brain signals during reading using 105-channel EEG
NEST Processing
Deep learning model processes signals through transformer layers
Text Output
Decoded thoughts appear as natural language text in real-time
Try It Yourself
Experience NEST's real-time EEG-to-text decoding with our interactive demo
EEG Signal Input
showing real-time brain activity
(105 channels, color-coded by region)
Decoded Text Output
"The quick brown fox jumps over the lazy dog..."
Powerful Features
Everything you need for brain signal decoding research and applications
Real-time Decoding
Process EEG signals and generate text output in milliseconds with our optimized inference pipeline.
State-of-the-Art Accuracy
Achieve 26.1% WER and 0.74 BLEU score on the ZuCo benchmark dataset.
Easy Integration
Simple Python API with pip installation. Get started in minutes with our comprehensive documentation.
Pre-trained Models
Download ready-to-use model checkpoints trained on ZuCo dataset for immediate deployment.
Research Ready
Full training pipeline for custom datasets. Fine-tune on your own EEG data with transfer learning.
Open Source
MIT licensed. Fully open-source codebase with active development and community support.
Start Decoding Brain Signals Today
Install NEST with pip and start decoding in minutes
# Install NEST pip install nest-eeg # Or install from source git clone https://github.com/wazder/NEST.git cd NEST pip install -e .