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 Experiment Video

Updated: Dec 6, 2025

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

882

Improving Speech Emotion Recognition With Adversarial Data Augmentation Network.

Lu Yi, Man-Wai Mak

    IEEE Transactions on Neural Networks and Learning Systems
    |October 9, 2020
    PubMed
    Summary
    This summary is machine-generated.

    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

    MicroRNA-125a-5p promotes NLRP3/caspase-1/GSDMD pathway-mediated pyroptosis in endothelial cells during Kawasaki disease.

    Scientific reports·2026
    Same author

    Correlation of foot process effacement with clinicopathological features and short-term prognosis in IgA nephropathy.

    Annals of medicine·2026
    Same author

    Efficacy and safety of telitacicept combined with immunosuppressive therapy for IgA nephropathy: a retrospective multicenter cohort study.

    Frontiers in immunology·2026
    Same author

    The predictive value of the pre-PCI prognostic nutritional index combined with the geriatric nutritional risk index for one-year outcomes in patients with chronic total occlusion.

    Frontiers in nutrition·2025
    Same author

    The differentiation between primary aldosteronism and essential hypertension in left ventricular remodeling before and after treatment: a follow-up study with quantitative stress cardiac magnetic resonance imaging.

    Quantitative imaging in medicine and surgery·2025
    Same author

    MutMap-Based Cloning of a Soybean Mosaic Virus Resistance Gene.

    Plants (Basel, Switzerland)·2025
    Same journal

    Granular Ball-Based Noise-Resistant Fuzzy Multineighborhood Feature Selection via Label Enhancement and Feature Graph.

    IEEE transactions on neural networks and learning systems·2026
    Same journal

    Fighting Evolving Spam With ARTMAP Models: A Noise-Resilient Online Detection Framework.

    IEEE transactions on neural networks and learning systems·2026
    Same journal

    HyperSAT: Unsupervised Hypergraph Neural Networks for Weighted MaxSAT Problems.

    IEEE transactions on neural networks and learning systems·2026
    Same journal

    Negation of Basic Belief Assignment in Multisource Information Fusion on Dempster-Shafer Theory With Applications in Pattern Classification.

    IEEE transactions on neural networks and learning systems·2026
    Same journal

    Intervention Feasible Region and Driver Risk Capacity Aware Human-Machine Collaborative Safe Trajectory Planning.

    IEEE transactions on neural networks and learning systems·2026
    Same journal

    A Unified Differential Denoising Learning Framework With a Pre-Trained Model and Fuzzy Graph Networks for Drug-Drug Interaction Prediction.

    IEEE transactions on neural networks and learning systems·2026
    See all related articles

    This study introduces the Adversarial Data Augmentation Network (ADAN) to combat overfitting in deep learning with limited data. ADAN generates synthetic data, enhancing speech emotion recognition accuracy.

    Area of Science:

    • Artificial Intelligence
    • Machine Learning
    • Deep Learning

    Background:

    • Deep neural networks (DNNs) are prone to overfitting when training data is scarce.
    • Data augmentation is crucial for improving DNN performance in low-data regimes.

    Purpose of the Study:

    • To propose a novel data augmentation network, the Adversarial Data Augmentation Network (ADAN), to address overfitting in DNNs.
    • To enhance the performance of speech emotion recognition systems using ADAN.

    Main Methods:

    • ADAN integrates a Generative Adversarial Network (GAN), an autoencoder, and an auxiliary classifier.
    • Class-dependent feature vectors are synthesized in latent and original feature spaces using Wasserstein divergence for adversarial training.
    • The method forces synthetic and real latent vectors to share a common representation to alleviate gradient vanishing.

    Related Experiment Videos

    Last Updated: Dec 6, 2025

    Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
    03:14

    Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

    Published on: December 6, 2024

    882

    Main Results:

    • ADAN effectively synthesizes high-quality, class-dependent feature vectors.
    • The synthesized data retains rich emotion information, improving classifier performance.
    • The approach significantly alleviates the gradient vanishing problem.

    Conclusions:

    • ADAN offers a robust solution for data augmentation in scenarios with limited training data.
    • The proposed method leads to competitive performance in speech emotion recognition.
    • ADAN demonstrates the potential of GAN-based approaches for enhancing deep learning models.