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

Large Language Model-Guided Design of Anti-Swelling Hybrid Dual Network Membranes for Long-Duration Alkaline Zinc Iron Flow Batteries.

Small (Weinheim an der Bergstrasse, Germany)·2026
Same author

Systematic design and analytical evaluation of a novel multi-speed bicycle rear transmission hub with a permanent magnet assisted gear shifting mechanism.

Scientific reports·2026
Same author

Luteolin Protects Against Noise-Induced Hearing Loss via Mitigating Oxidative Stress and Apoptosis, With Potential Regulation of the EGR1/SPRY4 Axis.

CNS neuroscience & therapeutics·2026
Same author

Aldose reductase inhibition suppresses microglial endoplasmic reticulum stress and ferritinophagy in ischemic stroke.

Acta pharmacologica Sinica·2026
Same author

DAMPs and PAMPs driven macrophage activation in kidney injury.

Tissue & cell·2026
Same author

Erchen decoction promotes inguinal white adipose tissue browning via the M2 macrophage-sympathetic axis to ameliorate obesity in mice.

Journal of ethnopharmacology·2026
Same journal

Amplitude-invariant phase masking for coherence recovery in scattered wavefields.

JASA express letters·2026
Same journal

Detecting continuous and discrete frequency changes as a function of spectral resolvability and modulation rate.

JASA express letters·2026
Same journal

Bearings-only acoustic source localization method using two distributed gliders and deep ocean experimental validation in the South China Sea.

JASA express letters·2026
Same journal

Block-sparse enhancement and detection of envelope modulation on noise for ship radiated noise.

JASA express letters·2026
Same journal

Predicting acoustic field with a separate variable ocean physics-informed neural network.

JASA express letters·2026
Same journal

Extending Sottek Hearing Model loudness to estimate partially-masked sound qualities of loudness, tonality, and sharpness.

JASA express letters·2026
See all related articles

Related Experiment Video

Updated: May 16, 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

466

Enhancing speech intelligibility in optical microphone systems through physics-informed data augmentation.

Jia-Wei Chen1, Jia-Hui Li1, Yi-Hao Jiang1

  • 1Department of Biomedical Engineering, National Yang Ming Chiao Tung University, Taipei, Taiwan.

JASA Express Letters
|April 2, 2025
PubMed
Summary
This summary is machine-generated.

This study introduces a data augmentation method to improve speech clarity from Laser Doppler Vibrometers (LDVs). The technique enhances intelligibility in noisy environments by simulating material distortions and noise for better audio processing.

More Related Videos

A Protocol for Comprehensive Assessment of Bulbar Dysfunction in Amyotrophic Lateral Sclerosis ALS
12:43

A Protocol for Comprehensive Assessment of Bulbar Dysfunction in Amyotrophic Lateral Sclerosis ALS

Published on: February 21, 2011

34.3K
Optical Scatter Microscopy Based on Two-Dimensional Gabor Filters
14:58

Optical Scatter Microscopy Based on Two-Dimensional Gabor Filters

Published on: June 2, 2010

9.5K

Related Experiment Videos

Last Updated: May 16, 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

466
A Protocol for Comprehensive Assessment of Bulbar Dysfunction in Amyotrophic Lateral Sclerosis ALS
12:43

A Protocol for Comprehensive Assessment of Bulbar Dysfunction in Amyotrophic Lateral Sclerosis ALS

Published on: February 21, 2011

34.3K
Optical Scatter Microscopy Based on Two-Dimensional Gabor Filters
14:58

Optical Scatter Microscopy Based on Two-Dimensional Gabor Filters

Published on: June 2, 2010

9.5K

Area of Science:

  • Acoustics and Signal Processing
  • Machine Learning for Audio Enhancement
  • Biomedical Engineering

Background:

  • Laser Doppler Vibrometers (LDVs) enable noncontact speech acquisition but suffer from material-dependent distortions and speckle noise.
  • These distortions significantly degrade speech intelligibility, especially in noisy environments.

Purpose of the Study:

  • To propose a novel data augmentation method to simulate and mitigate LDV-induced speech distortions.
  • To enhance speech intelligibility for LDV recordings under various material and low signal-to-noise ratio (SNR) conditions.

Main Methods:

  • Developed a data augmentation technique incorporating material-specific and impulse noises to mimic LDV distortions.
  • Employed a gated convolutional neural network integrated with HiFi-GAN for speech enhancement.

Main Results:

  • Achieved a Short-Time Objective Intelligibility (STOI) score of 0.76 at 0 dB SNR.
  • Demonstrated significant improvement in speech intelligibility across diverse materials and low SNR scenarios.

Conclusions:

  • The proposed data augmentation and deep learning approach effectively enhances speech intelligibility from LDV recordings.
  • Offers valuable insights for optimizing augmentation strategies and deep learning models in practical LDV-based audio applications.