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 Videos

A deformable convolution and attention-driven multiscale transformer framework for robust and explainable speech

Ati Jain1, Lalji Prasad1, Rashmi Yadav2

  • 1Institute of Advance Computing - Specialization, SAGE University, Indore, Madhya Pradesh, India.

Computer Methods in Biomechanics and Biomedical Engineering
|May 15, 2026
PubMed
Summary

Related Concept Videos

Labeling Emotion01:20

Labeling Emotion

Emotional labeling is a cognitive process that involves identifying and naming one's emotions, such as anger, fear, happiness, or sadness. It allows individuals to recognize and express their internal emotional states, a critical aspect of emotional regulation and communication. Labeling emotions requires more than mere recognition; it also involves drawing upon memory and contextual cues to understand the current situation and apply a corresponding emotional label. For instance, feeling...

You might also read

Related Articles

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

Sort by
Same author

Generational variation and stabilization in resynthesized allotetraploid Brassica juncea derived from diploid progenitors B. rapa and B. nigra.

BMC plant biology·2026
Same author

Base modifications shift tertiary structure and activity in synthetic RNA origami and a natural ribozyme.

Nature communications·2026
Same author

Multigene analysis reveals '<i>Candidatus</i> Phytoplasma asiaticum' (16SrII-C) association with niger and sesame phyllody in Madhya Pradesh, Central India and identification of a putative vector.

Frontiers in plant science·2026
Same author

Sex-specific differences in coronary blood flow dynamics during coronary artery bypass surgery.

Journal of cardiothoracic surgery·2026
Same author

The Heart Under Stress: What Happens When Both Ventricles Fail?

Echocardiography (Mount Kisco, N.Y.)·2026
Same author

Early-Onset Bilateral Clear Cell Renal Cell Carcinoma in a Pediatric Renal Transplant Recipient: A Rare Case Report.

Case reports in urology·2026
This summary is machine-generated.

This study presents DUST-Net, a novel network for robust Speech Emotion Recognition (SER). The advanced model achieves 98.75% accuracy, enhancing human-computer interaction and mental health applications.

Area of Science:

  • Artificial Intelligence
  • Human-Computer Interaction
  • Affective Computing

Background:

  • Speech Emotion Recognition (SER) is crucial for HCI, mental health diagnosis, and affective computing.
  • Existing SER methods face challenges due to speech signal non-stationarity and speaker variability.

Purpose of the Study:

  • To introduce a robust and efficient Speech Emotion Recognition (SER) system.
  • To enhance SER performance by addressing signal non-stationarity and speaker variability.

Main Methods:

  • Developed a deformable convolution-enhanced hierarchical Swin Transformer network (DUST-Net).
  • Integrated wavelet-based multiscale feature extraction, deformable convolutions, and adaptive token pruning.
  • Employed signal preprocessing, Hiking Optimization Algorithm optimization, and DeepLIFT interpretability.
Keywords:
Deformable convolutiondual attentionmultiscale transformerspeech emotion recognitiontoken pruningwavelet

Related Experiment Videos

Main Results:

  • Achieved a high accuracy of 98.75% in emotion recognition tasks.
  • Demonstrated robustness and generalizability of the DUST-Net model.
  • Showcased the efficiency of the proposed network in capturing emotional features.

Conclusions:

  • DUST-Net offers a significant advancement in robust and efficient Speech Emotion Recognition.
  • The model's interpretability enhances transparency in affective computing applications.
  • This research contributes to improved human-computer interaction and mental health diagnostics through advanced SER.