nm000224
Imagined speech EEG dataset — short and long words (Nguyen et al. 2017)
EEG
This dataset comprises preprocessed EEG recordings from 6 healthy participants performing imagined speech discrimination tasks between short and long words ('cooperate' vs 'in'). Data were acquired at 256 Hz using 64 EEG channels with standard preprocessing including bandpass filtering (8-70 Hz), notch filtering (60 Hz), and artifact removal. The dataset contains 1,200 trials analyzed using Riemannian manifold and relevance vector machine approaches for brain-computer interface applications, achieving mean classification accuracy of 73.3±8.9%.
- Size
- 236 MB
- Updated
- Jun 23, 2026