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

Robust acoustic object detection.

Yali Amit1, Alexey Koloydenko, Partha Niyogi

  • 1Departments of Computer Science and Statistics, The University of Chicago, Hyde Park, Chicago, Illinois 60637, USA. amit@galton.uchicago.edu

The Journal of the Acoustical Society of America
|November 4, 2005
PubMed
Summary
This summary is machine-generated.

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

Diagnostic accuracy of autofluorescence-Raman microspectroscopy for surgical margin assessment during Mohs micrographic surgery of basal cell carcinoma.

The British journal of dermatology·2024
Same author

MAP segmentation in Bayesian hidden Markov models: a case study.

Journal of applied statistics·2022
Same author

Biologically Plausible Training Mechanisms for Self-Supervised Learning in Deep Networks.

Frontiers in computational neuroscience·2022
Same author

Clinical integration of fast Raman spectroscopy for Mohs micrographic surgery of basal cell carcinoma.

Biomedical optics express·2021
Same author

Temperature-dependent mucosal permeation kinetics of stigmasterol microspheres: In vivo mice model antioral candidiasis study.

Journal of biomedical materials research. Part B, Applied biomaterials·2019
Same author

Deep Learning With Asymmetric Connections and Hebbian Updates.

Frontiers in computational neuroscience·2019
Same journal

Sibilant differentiation before and after tongue cancer surgery: Acoustics, kinematics and the role of sensorimotor controla).

The Journal of the Acoustical Society of America·2026
Same journal

BioNet-A: Ultrasonic echo representation network for target discrimination using active SONAR.

The Journal of the Acoustical Society of America·2026
Same journal

Empty soft-drink cans and mass-loaded rods: Analogous homework problems from acoustic and mechanical domains.

The Journal of the Acoustical Society of America·2026
Same journal

Erratum: Statistical wave field theory: Anisotropic wave fields under Neumann's boundary condition [J. Acoust. Soc. Am. 159(3), 2265-2280 (2026)].

The Journal of the Acoustical Society of America·2026
Same journal

On the modification of tip leakage noise sources by porous treatment.

The Journal of the Acoustical Society of America·2026
Same journal

An educational opportunity: Acoustics in an empty room.

The Journal of the Acoustical Society of America·2026
See all related articles

This study introduces a new method for speech recognition, detecting phonological units like words directly from audio. The approach uses robust time-frequency features to identify patterns, improving accuracy in noisy conditions.

Area of Science:

  • Speech processing
  • Phonetics
  • Signal analysis

Background:

  • Detecting phonological units (phonemes, syllables, words) directly from speech signals is challenging.
  • Existing methods may lack robustness to variations in intensity, time warping, noise, and competing speakers.

Purpose of the Study:

  • To present a novel approach for direct detection of phonological objects from the speech signal.
  • To develop robust and interpretable global templates for phonological units using local time-frequency features.

Main Methods:

  • Defining local features in the time-frequency domain with built-in robustness to intensity variations and time warping.
  • Constructing global templates based on the statistical coincidence of local feature patterns.
  • Evaluating diphone detectors, a word detector, and performing phonetic classification experiments.

Related Experiment Videos

Main Results:

  • The proposed global templates exhibit clear phonetic interpretability and adaptability.
  • The method demonstrates robustness against additive noise and competing speakers.
  • Performance evaluations show effectiveness for diphone detection, word detection, and phonetic classification.

Conclusions:

  • The novel approach offers a principled and robust method for detecting phonological objects directly from speech.
  • The developed templates provide phonetic interpretability and invariance, suitable for various speech recognition tasks.
  • This technique shows promise for improving the accuracy and reliability of speech processing systems.