IIST BCI Dataset-12 : A BCI Dataset for Telugu Vocal and Subvocal Commands with Dialect Variation for Wheelchair Control

Brain-computer interfaces (BCIs) have a lot of potential to assist people with motor disabilities in controlling various assistive devices using brain signals. However, the lack of datasets tailored to Indian languages, especially Telugu, a major language spoken by 81 million people, severely hampers its potential for native speakers. A novel electroencephalogram (EEG) based dataset, IIST BCI Dataset 12, is created using various Telugu words of various slangs and dialects. These words include commands such as Forward, Reverse, Left, Right, Go and Stop. The dataset was collected from two volunteers from different regions of Telangana and Andhrapradesh. The dataset aims to aid Machine Learning (ML) and BCI research, enabling significant contribution for adaptive and accurate EEGbased wheelchair control systems for native Telugu speakers.