TISE Dataset-Thai Script for Imaginary Spelling using Electroencephalogram Signals

Description

Thai Script for Imaginary Spelling using Electroencephalogram (TISE dataset) is related to pronunciation/spelling and imagination of Thai vowel script based on electroencephalogram (EEG) signals. This dataset consists of character spelling (CS) and character imagination (CI) datasets. Providing CS and CI datasets is very important to ensure that the EEG signal patterns of spoken scripts are by imagination. So, this dataset needs to be provided. In addition, previous researchers have yet to collect Thai script imagination datasets. The participants used as samples in this study were 33 healthy people, 25 males, and eight females, from Thai students at the Rajamangala University of Technology Thanyabur. The participants' EEG signals were recorded using Contec KT 88 with 16 channels. This dataset consists of 4 types of data: (1) Raw data for experiments, (2) Raw data for calm conditions, (3) Data analysis for Character spelling, and (4) Data analysis for Character imagination. Data analysis for Character spelling and Character imagination was generated from the combined data (joined_data) for eight Thai vowel characters based on the timestamp of the characters' appearance. Also attached is the Ethical Clearance.pdf file related to the research permit from the Ganesha University of Education ethics committee. The Stimulus media.mp4 file is the stimulus media file used during the experiment, while the Timestamp Thai script.pdf file contains the duration of the Thai characters' appearance.

Steps to reproduce

a. Preparation of research documents Some documents that must be prepared to collect EEG signal data in this study include: 1) Research information documents, 2) Consent Form, 3) Preparation of health information b. Preparation of stimulus media This study will use two types of stimulus media, namely Thai script stimulus media. This stimulus media is in the form of alphabet images containing speech sounds. Each participant will be asked to pronounce all Thai letters. Through this scenario, an EEG signal pattern related to Thai letters can be produced c. Submission of ethical permission To ensure that the recording process runs well and smoothly and minimizes its negative impact on participants, the research team proposes ethical permission for this recording. The following is the ethical permission that has been approved by the Faculty of Medicine, Universitas Pendidikan Ganesha d. Determine participants The procedure for determining participants is based on research conducted by Koelstra et al, Katsigiannis and Ramzan, and research conducted by Miranda Correa et al. e. Recording process This scenario is conducted in 2 cycles for each participant, imagining and pronouncing Thai characters that appear sequentially and randomly. Each participant in each scenario will practice spelling and imagining Thai letters based on the stimulus media provided.

Institutions

Categories

Funding

Licence