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

Hearing01:31

Hearing

56.1K
When we hear a sound, our nervous system is detecting sound waves—pressure waves of mechanical energy traveling through a medium. The frequency of the wave is perceived as pitch, while the amplitude is perceived as loudness.
56.1K

You might also read

Related Articles

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

Sort by
Same author

Robust Three-Microphone Speech Source Localization Using Randomized Singular Value Decomposition.

IEEE access : practical innovations, open solutions·2021
Same author

Smartphone-based single-channel speech enhancement application for hearing aids.

The Journal of the Acoustical Society of America·2021
Same author

Spectral Flux-Based Convolutional Neural Network Architecture for Speech Source Localization and Its Real-Time Implementation.

IEEE access : practical innovations, open solutions·2021
Same author

CONVOLUTIONAL RECURRENT NEURAL NETWORK BASED DIRECTION OF ARRIVAL ESTIMATION METHOD USING TWO MICROPHONES FOR HEARING STUDIES.

IEEE International Workshop on Machine Learning for Signal Processing : [proceedings]. IEEE International Workshop on Machine Learning for Signal Processing·2021
Same author

Real-time single-channel deep neural network-based speech enhancement on edge devices.

Interspeech·2021
Same author

Behavioral Validation of the Smartphone for Remote Microphone Technology.

Seminars in hearing·2020

Related Experiment Video

Updated: Dec 13, 2025

Author Spotlight: Optimizing EAS with Long Electrodes for Enhanced Cochlear Coverage and Hearing Preservation
03:49

Author Spotlight: Optimizing EAS with Long Electrodes for Enhanced Cochlear Coverage and Hearing Preservation

Published on: October 11, 2024

1.1K

Efficient two-microphone speech enhancement using basic recurrent neural network cell for hearing and hearing aids.

Nikhil Shankar1, Gautam Shreedhar Bhat1, Issa M S Panahi1

  • 1Department of Electrical and Computer Engineering, The University of Texas at Dallas, Richardson, Texas 75080, USA.

The Journal of the Acoustical Society of America
|August 6, 2020
PubMed
Summary

This study introduces a real-time speech enhancement (SE) system using a recurrent neural network (RNN) and two microphones. The efficient framework improves speech quality and intelligibility for hearing aid devices (HADs) in noisy conditions.

More Related Videos

Systematic Hearing Performance Evaluation Process for Adolescents with Cochlear Implantation at Early Ages
06:04

Systematic Hearing Performance Evaluation Process for Adolescents with Cochlear Implantation at Early Ages

Published on: March 24, 2023

671
Neuro-rehabilitation Approach for Sudden Sensorineural Hearing Loss
09:44

Neuro-rehabilitation Approach for Sudden Sensorineural Hearing Loss

Published on: January 25, 2016

19.6K

Related Experiment Videos

Last Updated: Dec 13, 2025

Author Spotlight: Optimizing EAS with Long Electrodes for Enhanced Cochlear Coverage and Hearing Preservation
03:49

Author Spotlight: Optimizing EAS with Long Electrodes for Enhanced Cochlear Coverage and Hearing Preservation

Published on: October 11, 2024

1.1K
Systematic Hearing Performance Evaluation Process for Adolescents with Cochlear Implantation at Early Ages
06:04

Systematic Hearing Performance Evaluation Process for Adolescents with Cochlear Implantation at Early Ages

Published on: March 24, 2023

671
Neuro-rehabilitation Approach for Sudden Sensorineural Hearing Loss
09:44

Neuro-rehabilitation Approach for Sudden Sensorineural Hearing Loss

Published on: January 25, 2016

19.6K

Area of Science:

  • Speech Processing
  • Artificial Intelligence
  • Signal Processing

Background:

  • Noisy environments degrade speech quality and intelligibility.
  • Existing speech enhancement (SE) methods have limitations in real-time performance and effectiveness at low signal-to-noise ratios (SNRs).
  • Hearing aid devices (HADs) require robust SE solutions for improved user experience.

Purpose of the Study:

  • To develop a real-time, two-microphone speech enhancement (SE) framework using a recurrent neural network (RNN).
  • To improve speech quality and intelligibility in noisy environments, particularly for hearing aid devices (HADs).
  • To demonstrate the computational efficiency and minimal processing delay of the proposed SE algorithm.

Main Methods:

  • A basic recurrent neural network (RNN) cell was employed for speech enhancement.
  • The RNN model was trained using the real and imaginary parts of the short-time Fourier transform (STFT) as features.
  • The algorithm was implemented in real-time on a smartphone platform utilizing its two built-in microphones.

Main Results:

  • The proposed RNN-based SE method demonstrated superior performance compared to conventional SE techniques.
  • Significant improvements in speech quality and intelligibility were observed across various noise conditions and low SNRs.
  • The real-time implementation on a smartphone showed computational efficiency and minimal input-output delay.

Conclusions:

  • The developed two-microphone SE framework offers an effective and efficient solution for enhancing speech in noisy environments.
  • The RNN-based approach shows promise as an assistive tool for hearing aid devices (HADs).
  • The method's real-time capability and performance at low SNRs make it suitable for mobile and wearable applications.