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

Effects of game interactivity on perceived emotional challenge, mixed-affect emotional experiences and physiological responses.

Frontiers in psychology·2026
Same author

Research on the Robustness of Focus Measure Operators Based on RRMSE.

Sensors (Basel, Switzerland)·2026
Same author

From stiffness-softness paradox to drug delivery systems: mechanical properties and regulation in tumor progression.

International journal of pharmaceutics·2026
Same author

Formulation Design of a Hot-melt Extrusion-Based Amorphous Solid Dispersion to Optimize Palatability of Berberine Hydrochloride with In Vitro and In Vivo Evaluation.

AAPS PharmSciTech·2026
Same author

Population history and subsistence of farming communities in an agro-pastoral transition zone of northern China: ancient DNA and isotopic evidence from the Erdaojingzi site.

Scientific reports·2026
Same author

Mechanisms of heat and hypoxia defense in the sea cucumber Apostichopus japonicus: Insights from ubiquitination regulation.

Comparative biochemistry and physiology. Part D, Genomics & proteomics·2026
Same journal

Consensus by design: Conflict of interest and the intravenous ketamine Delphi guidelines.

Journal of affective disorders·2026
Same journal

Impaired crying despite preserved emotional experience under antidepressant treatment: An under-recognized clinical phenomenon.

Journal of affective disorders·2026
Same journal

Corrigendum to "Effects of neuronavigation-guided rTMS on serum BDNF, TrkB and VGF levels in depressive patients with suicidal ideation" [J. Affect. Disord. 323 (2023) 617-623].

Journal of affective disorders·2026
Same journal

Transcutaneous vagus nerve stimulation enhances reward-effort efficiency in severe major depressive disorder.

Journal of affective disorders·2026
Same journal

Age-varying associations between attitudes toward suicide and suicidal ideation in Chinese psychiatric outpatients.

Journal of affective disorders·2026
Same journal

Aripiprazole once-monthly for patients diagnosed with bipolar I disorder: Number needed to treat, number needed to harm, and likelihood to be helped or harmed.

Journal of affective disorders·2026
See all related articles

Related Experiment Video

Updated: Jan 17, 2026

Asthma Detection Research Based on Voice Signal Processing and Machine Learning
04:04

Asthma Detection Research Based on Voice Signal Processing and Machine Learning

Published on: July 22, 2025

934

Auto-Masked Audio Spectrogram Transformer for depression detection from speech.

Mianchen Zhang1, Jian He2, Xiaolan Peng3

  • 1College of Computer Science and Technology, Beijing University of Technology, No. 100, Pingleyuan, Beijing 100124, China.

Journal of Affective Disorders
|September 18, 2025
PubMed
Summary
This summary is machine-generated.

This study presents the Auto-Masked Audio Spectrogram Transformer (AMAST), a novel deep learning model for depression detection using speech. AMAST achieves high accuracy in identifying depressive speech markers, offering a non-invasive screening tool.

Keywords:
Audio spectrogram transformerAuto-maskedMajor depression disorderSliding windowTime–frequency attention

More Related Videos

Author Spotlight: Therapeutic Benefit of Closed-Loop Deep Brain Stimulation in Depression Treatment
05:19

Author Spotlight: Therapeutic Benefit of Closed-Loop Deep Brain Stimulation in Depression Treatment

Published on: July 7, 2023

3.3K
Computer-based Multitaper Spectrogram Program for Electroencephalographic Data
04:13

Computer-based Multitaper Spectrogram Program for Electroencephalographic Data

Published on: November 13, 2019

12.8K

Related Experiment Videos

Last Updated: Jan 17, 2026

Asthma Detection Research Based on Voice Signal Processing and Machine Learning
04:04

Asthma Detection Research Based on Voice Signal Processing and Machine Learning

Published on: July 22, 2025

934
Author Spotlight: Therapeutic Benefit of Closed-Loop Deep Brain Stimulation in Depression Treatment
05:19

Author Spotlight: Therapeutic Benefit of Closed-Loop Deep Brain Stimulation in Depression Treatment

Published on: July 7, 2023

3.3K
Computer-based Multitaper Spectrogram Program for Electroencephalographic Data
04:13

Computer-based Multitaper Spectrogram Program for Electroencephalographic Data

Published on: November 13, 2019

12.8K

Area of Science:

  • Artificial Intelligence
  • Computational Linguistics
  • Psychiatry

Background:

  • Depression is a psychological disorder marked by cognitive and emotional changes.
  • Current diagnostic methods for depression can be expensive and invasive.
  • Speech analysis presents a viable, accessible method for early depression detection.

Purpose of the Study:

  • To introduce the Auto-Masked Audio Spectrogram Transformer (AMAST) for depression detection from speech.
  • To evaluate AMAST's performance in extracting depression-related features from speech spectrograms.
  • To provide an accessible, non-invasive tool for mental health assessment.

Main Methods:

  • The Auto-Masked Audio Spectrogram Transformer (AMAST) deep learning framework was developed.
  • AMAST utilizes sliding window segmentation and auto-masked training for enhanced contextual learning.
  • A time-frequency attention mechanism was incorporated to capture comprehensive speech information.

Main Results:

  • AMAST achieved high F1 scores (0.92 and 0.91) on two distinct datasets.
  • Performance was significantly enhanced by emotionally evocative speech tasks like interviews.
  • The model demonstrated robustness in detecting subtle depressive speech markers under various conditions.

Conclusions:

  • AMAST shows significant promise as a non-invasive tool for depression screening.
  • The model's effectiveness across diverse tasks and datasets supports its clinical utility.
  • AMAST can aid in remote mental health assessments and early intervention strategies.