ArEEG: an Open-Access Arabic Inner Speech EEG Dataset.

Recent advancements in Brain-Computer Interface (BCI) technology are shifting towards inner speech over motor imagery due to its intuitive nature and broader command spectrum, enhancing interaction with electronic devices. However, the reliance on a large number of electrodes in available datasets complicates the development of cost-effective BCIs. Additionally, the lack of publicly available datasets hinder the development of this technology. To address this, we introduce a new Arabic Inner Speech dataset, featuring five distinct classes, exceeding the typical four-class datasets, and recorded using only eight electrodes, making it an economical solution. Our primary objective is to provide an open-access, multi-class Electroencephalographic (EEG) dataset in Arabic for inner speech, encompassing five commands. This dataset is designed to enhance our understanding of brain activity, facilitate the integration of BCI technologies in Arabic-speaking regions, and serve as a valuable resource for developing and testing real-world BCI applications. Through this contribution, we aim to bridge the gap between language-specific neural data and the field of neurotechnology, fostering innovation and inclusivity in BCI research.