TISE Dataset-Thai Script for Imaginary Spelling using Electroencephalogram Signals

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.基于脑电图的泰语假想拼写数据集（Thai Script for Imaginary Spelling using Electroencephalogram，TISE dataset）围绕基于脑电图（Electroencephalogram，EEG）信号的泰语元音文字的发音、拼写与想象任务构建。该数据集包含字符拼写（Character Spelling，以下简称CS）数据集与字符想象（Character Imagination，以下简称CI）数据集两类子数据集。构建CS与CI数据集的核心意义在于，确保所采集的语音文字相关脑电图信号模式均源于想象过程，因此本数据集的构建十分必要。此外，此前尚无研究者开展泰语文字想象相关的数据集采集工作。本研究的受试样本为33名健康个体，其中男性25名、女性8名，均为泰国吞武里皇家理工大学（Rajamangala University of Technology Thanyabur）的在校学生。研究采用康泰KT 88（Contec KT 88）16通道设备采集受试者的脑电图信号。该数据集包含四类数据：(1) 实验原始数据；(2) 静息状态原始数据；(3) 字符拼写任务分析数据；(4) 字符想象任务分析数据。字符拼写与字符想象任务的分析数据，基于8个泰语元音字符的呈现时间戳，由合并后的联合数据（joined_data）生成。随数据集一并附上的还有由甘尼萨教育大学（Ganesha University of Education）伦理委员会出具的研究许可伦理审批文件（Ethical Clearance.pdf）。实验刺激媒体文件（Stimulus media.mp4）为实验过程中使用的刺激素材文件，而泰语字符时间戳文件（Timestamp Thai script.pdf）则记录了各泰语字符的呈现时长。