emg2qwerty: A Large Dataset with Baselines for Touch Typing using Surface Electromyography
NEMAR Dataset nm000104: emg2qwerty - EMG-based typing detection
- Participants
- 108
- Size
- 223 GB
- Version
- vv2.0.0
- Updated
- Jul 10, 2026
15 results for "Handwriting" · page 2 of 2 · ranked by relevance
NEMAR Dataset nm000104: emg2qwerty - EMG-based typing detection
EEG recordings from 10 healthy participants using an auditory P300 brain-computer interface speller based on the AMUSE (Auditory Multi-class Spatial ERP) paradigm. Six speakers placed around the participant provided spatial auditory cues for a two-step, hex-o-spell-like character selection. The recordings include calibration and online spelling in two sessions, the second using a dynamic stopping method to increase writing speed.
A left/right-hand motor-imagery EEG dataset from 29 healthy participants recorded at the University of West Bohemia for brain-computer interface research. Participants performed cue-based single-class imagery of left- or right-hand movement, with four runs per session. The data were recorded with BrainVision hardware and re-released in BIDS format through MOABB, with HED event annotations.
A collection of electroencephalography (EEG) recordings from 15 students (8 male, 7 female, aged 21-23) performing essay-writing tasks. The task was designed to assess cognitive conditions such as creativity, memory, and critical thinking under different generative AI interaction conditions. The dataset supports studies of brain connectivity during student learning processes with Gen-AI.
EEG recordings from 13 healthy participants performing an online copy-spelling task with a visual 6x7 matrix ERP speller. The speller was designed for learning from label proportions (LLP), an unsupervised, calibration-free classification approach for ERP-based brain-computer interfaces. Two stimulus sequences with different target/non-target ratios allowed the classifier to reconstruct mean target and non-target responses without labeled data. Participants spelled a German pangram three times, with 84.5% of characters spelled correctly on average without prior calibration.