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

An artificial neural network for sound localization using binaural cues

M S Datum1, F Palmieri, A Moiseff

  • 1Engineering Technology Center, Mystic, Connecticut 06355-1208, USA.

The Journal of the Acoustical Society of America
|July 1, 1996
PubMed
Summary

A novel three-layer neural network accurately estimates sound source direction using two receivers. Trained with simulated data and the MEKA algorithm, it offers efficient and reliable acoustic localization.

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

Robotic ventral mesh rectopexy for obstructed defecation syndrome using the Hugo RAS™ system-A video vignette.

Colorectal disease : the official journal of the Association of Coloproctology of Great Britain and Ireland·2025
Same author

Early presentation of Covid-19 related paracentral acute middle maculopathy in a healthy young patient.

Journal francais d'ophtalmologie·2024
Same author

Drugs for treating infections caused by non-tubercular mycobacteria: a narrative review from the study group on mycobacteria of the Italian Society of Infectious Diseases and Tropical Medicine.

Infection·2024
Same author

VIDAS<sup>®</sup> TB-IGRA reagents induce a CD4<sup>+</sup> and CD8<sup>+</sup> T-cell IFN-γ response for both TB infection and active TB.

The international journal of tuberculosis and lung disease : the official journal of the International Union against Tuberculosis and Lung Disease·2021
Same author

Therapeutic strategies and outcomes of MDR and pre-XDR-TB in Italy: a nationwide study.

The international journal of tuberculosis and lung disease : the official journal of the International Union against Tuberculosis and Lung Disease·2021
Same author

Outcome of treatment of MDR-TB or drug-resistant patients treated with bedaquiline and delamanid: Results from a large global cohort.

Pulmonology·2021

Area of Science:

  • Acoustics
  • Artificial Intelligence
  • Signal Processing

Background:

  • Directional sound source localization is crucial in various applications.
  • Traditional methods often require detailed environmental knowledge or complex calibration.
  • Developing robust and adaptable localization systems remains an active research area.

Purpose of the Study:

  • To develop and evaluate a neural network-based system for estimating sound source direction.
  • To investigate the effectiveness of the multiple extended Kalman algorithm (MEKA) for training such networks.
  • To compare the performance of the neural network with theoretical lower bounds on estimation accuracy.

Main Methods:

  • Utilized a three-layer neural network architecture.
  • Employed simulated acoustical environment data for network training.

Related Experiment Videos

  • Implemented the multiple extended Kalman algorithm (MEKA) for efficient network training.
  • Computed and analyzed estimation lower bounds.

    • Main Results:

    • The neural network successfully estimated sound source direction from two spatially separated receivers.
    • The MEKA algorithm facilitated fast convergence during network training without parameter tuning.
    • Simulations demonstrated the network's performance, showing promising results compared to theoretical limits.

    Conclusions:

    • A three-layer neural network trained with MEKA is an effective method for sound source direction estimation.
    • The system demonstrates robustness, not requiring prior knowledge of acoustic parameters.
    • The approach shows potential for real-world acoustic localization applications.