Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Dynamic wavelet-based augmentation for enhanced EEG-based imagined speech classification.

Computers in biology and medicine·2026
Same author

Microbiota-derived indole derivatives as anticancer agents: mechanistic insights and major perspectives.

Future microbiology·2026
Same author

Microplastics in simulated digestion: Surface modifications, enzyme interference, and chemical migration.

Food chemistry·2026
Same author

Optimization of polyhydroxyalkanoate biopolymer production from lignocellulosic wood waste using statistical experimental designs.

Scientific reports·2026
Same author

Emerging implications of bacterial biofilm in cancer biology: Recent updates and major perspectives.

Gut microbes reports·2026
Same author

A Cloud-Aware Scalable Architecture for Distributed Edge-Enabled BCI Biosensor System.

Biosensors·2026
Same journal

HF-SNVTA-FusionNet: high-frequency multi-domain EEG feature fusion from the substantia nigra and ventral tegmental area for Parkinson's disease classification.

Cognitive neurodynamics·2026
Same journal

Investigation of the effects of balance exercises on visuospatial skills using EEG brain oscillations.

Cognitive neurodynamics·2026
Same journal

MSCANet: a cross-attention-based multi-scale convolutional fusion neural network for EEG motor imagery classification.

Cognitive neurodynamics·2026
Same journal

Regulation of epileptiform discharges in thalamocortical model based on preview control theory.

Cognitive neurodynamics·2026
Same journal

Computational modeling of tyrosine hydroxylase pathway for dopamine synthesis in nerve cells: effect of tetrahydrobiopterin deficiency and oxidative stress.

Cognitive neurodynamics·2026
Same journal

From nonlinear neuronal dynamics to AI-optimized VLSI hardware: multiplier-free FPGA implementation of memristive FN-HR coupled neural networks for intelligent systems.

Cognitive neurodynamics·2026
See all related articles

Related Experiment Video

Updated: Jan 16, 2026

Assessment and Communication for People with Disorders of Consciousness
07:37

Assessment and Communication for People with Disorders of Consciousness

Published on: August 1, 2017

9.6K

Innovative augmentation techniques and optimized ANN model for imagined speech decoding in EEG-based BCI.

Anand Mohan1, R S Anand1

  • 1Department of Electrical Engineering, Indian Institute of Technology Roorkee, Roorkee, Uttarakhand 247667 India.

Cognitive Neurodynamics
|September 30, 2025
PubMed
Summary
This summary is machine-generated.

This study enhances brain-computer interfaces (BCIs) for imagined speech decoding by using electroencephalogram (EEG) data augmentation. Gaussian noise augmentation achieved 91% accuracy for long words, improving communication for those with impairments.

Keywords:
AugmentationCovert speechEEGImagined speechMachine learningSignal processing

More Related Videos

A Single-Channel and Non-Invasive Wearable Brain-Computer Interface for Industry and Healthcare
06:34

A Single-Channel and Non-Invasive Wearable Brain-Computer Interface for Industry and Healthcare

Published on: July 7, 2023

3.1K
A Novel Experimental and Analytical Approach to the Multimodal Neural Decoding of Intent During Social Interaction in Freely-behaving Human Infants
11:14

A Novel Experimental and Analytical Approach to the Multimodal Neural Decoding of Intent During Social Interaction in Freely-behaving Human Infants

Published on: October 4, 2015

11.4K

Related Experiment Videos

Last Updated: Jan 16, 2026

Assessment and Communication for People with Disorders of Consciousness
07:37

Assessment and Communication for People with Disorders of Consciousness

Published on: August 1, 2017

9.6K
A Single-Channel and Non-Invasive Wearable Brain-Computer Interface for Industry and Healthcare
06:34

A Single-Channel and Non-Invasive Wearable Brain-Computer Interface for Industry and Healthcare

Published on: July 7, 2023

3.1K
A Novel Experimental and Analytical Approach to the Multimodal Neural Decoding of Intent During Social Interaction in Freely-behaving Human Infants
11:14

A Novel Experimental and Analytical Approach to the Multimodal Neural Decoding of Intent During Social Interaction in Freely-behaving Human Infants

Published on: October 4, 2015

11.4K

Area of Science:

  • Neuroscience and Rehabilitation Engineering
  • Brain-Computer Interface (BCI) Technology
  • Signal Processing and Machine Learning

Background:

  • Electroencephalogram (EEG) signals are crucial for Brain-Computer Interfaces (BCIs), particularly for decoding imagined speech in individuals with neurological impairments.
  • The nonstationary nature of EEG signals and limited dataset sizes pose significant challenges for developing robust machine learning algorithms.
  • Effective decoding of imagined speech via EEG-BCIs can empower communication for individuals with severe motor disabilities.

Purpose of the Study:

  • To propose and evaluate EEG data augmentation methods to address data scarcity and enhance the robustness of machine learning models for imagined speech decoding.
  • To introduce a novel neural network architecture designed to detect subtle variations in EEG signals specific to imagined speech tasks.
  • To demonstrate the effectiveness of proposed augmentation techniques and the novel architecture in improving classification performance compared to baseline methods.

Main Methods:

  • Investigated seven diverse EEG data augmentation techniques, including Gaussian noise, to increase dataset variability.
  • Developed and implemented a novel neural network architecture for analyzing EEG signal variations during imagined speech.
  • Evaluated model performance using accuracy, F1-score, and kappa metrics, comparing augmented data results against a non-augmented baseline.

Main Results:

  • The proposed model achieved a remarkable accuracy of 91% for classifying long words when utilizing Gaussian noise augmentation.
  • EEG data augmentation significantly improved model robustness and classification performance across various metrics.
  • The novel architecture demonstrated effectiveness in detecting discriminative features within augmented EEG datasets.

Conclusions:

  • EEG data augmentation is a viable strategy to overcome data scarcity challenges in Brain-Computer Interface (BCI) research for imagined speech.
  • The proposed novel architecture and augmentation methods show significant promise for advancing EEG-based communication systems.
  • This approach holds potential for restoring communication capabilities for individuals with speech and motor impairments.