Download Report IEEE DataPort : A Multi-Class Electroencephalography Dataset for Imagined Speech Decoding. - 2026
MAT, MATLAB by R.S. Anand, Meenakshi Bisla, Dehradun, Anand Mohan, Dilnawaz Dilnawaz
Information
Format: MAT, MATLAB Publisher: IEEE DataPort Publication Date of the Electronic Edition: 02/05/2026
?
ISBN: 10.21227/2ybw-rt36
$15
$3Discount Coupon
Delivery time: Maximum 24 hours
Description
This dataset contains 32-channel electroencephalography (EEG) recordings acquired for multi-class imagined speech decoding using a structured and controlled experimental protocol.Data were collected from ten male participants at the Biomedical Instrumentation Laboratory, Indian Institute of Technology (IIT) Roorkee.Following standard preparation proceduresincluding informed consent, EEG cap placement, and system calibrationparticipants performed tasks across three linguistic categories: vowels, words, and sentences.The signals were recorded using an Emotive Epoch Flex system at a sampling frequency of 128 Hz.The experimental design involved eight distinct stimuli: the vowels "a" and "i”; the words "water," "sleep," "fine," and "problem”; and the sentences "I am fine." and "I have problem.".Each trial followed a fixed temporal structure consisting of a 3-second stimuluspresentation, a 4-second imagined speech intervalwhere participants silently imagined articulating the stimulusand a 2-second relaxation phase.To ensure a robust sample size for machine learning and deep learning models, each stimulus was repeated across 60 trials per participant.The raw signals were preprocessed with a 50 Hz notch filter and a 0.1–64 Hz bandpass filter to mitigate noise while preserving physiological intent.The dataset is organized hierarchically and provided in MATLAB (.mat) format, enabling reproducible research in signal processing, machine learning, and brain–computer interface (BCI) studies focused on imagined speech.
$15
$3Discount Coupon
Delivery time: Maximum 24 hours