Thinking out loud, an open-access EEG-based BCI dataset for inner speech recognition

Area of Science:

• Neuroscience
• Artificial Intelligence
• Biomedical Engineering

Background:

• Surface electroencephalography (EEG) noninvasively measures brain activity.
• Artificial intelligence (AI) has improved Brain-Computer Interfaces (BCIs).
• Interpreting the 'inner voice' or inner speech phenomenon is a growing research area.

Purpose of the Study:

• To address the lack of publicly available EEG datasets for inner speech recognition.
• To provide an open-access, multiclass EEG database of inner speech commands.
• To facilitate the development of new techniques for inner speech recognition and understanding brain mechanisms.

Main Methods:

• Acquisition of a ten-participant dataset using a 136-channel EEG system.
• Recording data across inner speech and two related paradigms.
• Development of an open-access database for research purposes.

Main Results:

• A novel, publicly available EEG dataset specifically for inner speech commands has been created.
• The dataset comprises recordings from ten participants across multiple related paradigms.
• This resource supports the advancement of AI-driven BCIs for thought-based control.

Conclusions:

• The presented EEG dataset is a valuable resource for the scientific community.
• It will aid in the development of more accurate and accessible inner speech recognition systems.
• This contributes to a better understanding of the neural mechanisms underlying inner speech and its application in BCIs.