EEG-Speech Brain Decoding Dataset

# EEG-Speech Brain Decoding Dataset ## Overview This dataset contains EEG recordings and audio data. ## Sessions Sessions are labeled by recording date in YYYYMMDD format. - Example: `ses-20240401` = recorded on April 1, 2024 Multiple recordings on the same day are distinguished by run numbers: - `run-N`: Nth recording of the day ## Tasks - **speechopen**: Overt speech production task - Participants vocalize visually presented text - **listening**: Auditory listening task - Participants listen to prerecorded speech stimuli - **listeningcovert**: Auditory listening followed by covert speech imagery ## EEG Acquisition Devices Recordings are split by acquisition label: - **acq-pangolin**: g.tec g.Pangolin. Stored channels include 128 EEG channels plus audio monitor, EOG, EMG, trigger, and in 140-channel recordings one auxiliary mastoid/reference channel that is not used for analysis. - **acq-scarabeo**: g.tec g.SCARABEO. Stored channels include 64 EEG-related channels, with channels 62 and 63 corresponding to mastoid electrodes, plus audio monitor, EOG, EMG, and trigger channels. - **acq-eego**: ANT Neuro eego sports. Stored channels include EEG channels 0..30 and 32..63, channel 31 as miscellaneous, audio monitor, EOG, upper- and lower-lip EMG, miscellaneous auxiliary channels, and trigger. Run-level `*_channels.tsv` files provide the authoritative channel typing for each recording. For g.Pangolin sessions, `*_electrodes.tsv` and `*_coordsystem.json` files provide 3D coordinates for EEG001..EEG128 from g.tec electrode digitization (`electrodes_uhd.xml`). Electrode coordinate files are not provided for g.SCARABEO or eego sports sessions because verified channel-to-position coordinate mappings are not available in this release. ## File Format Notes ### EEG Data Raw EEG data is stored: - **Path**: `sub-*/ses-*/eeg/*_eeg.edf` - **Note**: EDF files are included as the raw EEG recordings for this BIDS-EEG dataset. ### Behavioral Data (Audio) Task audio files are stored in `beh/` directories: - **Speech production**: `sub-*/ses-*/beh/*_recording-vocal_beh.wav` - **Listening**: `sub-*/ses-*/beh/*_recording-audio_beh.wav` - **Note**: Not officially part of BIDS-EEG spec, but included for analysis convenience - Excluded in `.bidsignore` ## Directory Structure ``` dataset_root/ ├── README (this file) ├── CHANGES (version history) ├── dataset_description.json (dataset metadata) ├── participants.tsv (participant information) ├── participants.json (participant column descriptions) ├── task-speechopen_acq-pangolin_eeg.json (speech production EEG metadata) ├── task-speechopen_acq-scarabeo_eeg.json (speech production EEG metadata) ├── task-speechopen_acq-eego_eeg.json (speech production EEG metadata) ├── task-listening_acq-pangolin_eeg.json (listening EEG metadata) ├── task-speechopen_acq-pangolin_events.json (speech production events column descriptions) ├── task-speechopen_acq-pangolin_recording-vocal_beh.json ├── task-listening_acq-pangolin_recording-audio_beh.json ├── .bidsignore (files to ignore in validation) │ ├── code/ (analysis and preprocessing code) │ ├── preprocessing/ (EEG and audio preprocessing) │ ├── training/ (model training scripts) │ ├── evaluation/ (evaluation metrics) │ └── bids/ (BIDS conversion scripts) │ ├── sub-01/ (participant data) │ └── ses-YYYYMMDD/ (session by date) │ ├── eeg/ (EEG recordings) │ └── beh/ (behavioral/audio data) │ └── derivatives/ (processed data) └── pipeline-standard/ (standard preprocessing) ```