Inner Speech Dataset

Important! On 30 July 2021, several corrupted files were fixed in the data repository. On 25 November 2021, EEG data for participants 9 and 10 were also fixed in the repository.

In the following repository, all codes for reproducing and using the Inner speech Dataset are presented.

The dataset is publicly available at https://openneuro.org/datasets/ds003626

The publication is available at https://www.nature.com/articles/s41597-022-01147-2

Stimulation Protocol

The stimulation protocol was used for capturing the data and was developed in Matlab using Psychtoolbox.

The script Stimulation_protocol.m is the main script and uses the other auxiliary functions.

Processing

The processing was developed in Python, using mainly the MNE library.

Install the Inner speech processing environment

Create an environment with all the necessary libraries for running all the scripts.

conda env create -f environment.yml

Using the Inner_speech_processing.py script, you can easily make your processing, by changing the variables at the top of the script.

The TFR_representation.py generates the Time-Frequency Representations used addressing the same processing followed in the paper.

Using the Plot_TFR_Topomap.py the same images presented in the paper can be addressed.

Citing this work

Please cite this work.

@article{nieto2022thinking,
  title={Thinking out loud, an open-access EEG-based BCI dataset for inner speech recognition},
  author={Nieto, Nicol{\'a}s and Peterson, Victoria and Rufiner, Hugo Leonardo and Kamienkowski, Juan Esteban and Spies, Ruben},
  journal={Scientific Data},
  volume={9},
  number={1},
  pages={1--17},
  year={2022},
  publisher={Nature Publishing Group}
}

@article{nieto2021inner,
  title={Inner Speech},
  author={Nieto, N and Peterson, V and Rufiner, HL and Kamienkowski, JE and Spies, R},
  journal={OpenNeuro},
  volume={29},
  pages={227--236},
  year={2021}
}