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

Cochannel speech separation.

C K Lee1, D G Childers

  • 1Department of Electrical Engineering, University of Florida, Gainesville 32611.

The Journal of the Acoustical Society of America
|January 1, 1988
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

A study of the predonation hemoglobin and iron status among Hong Kong Chinese blood donors.

Transfusion·2012
Same author

Effectiveness and cost-effectiveness of sentinel lymph node biopsy compared with axillary node dissection in patients with early-stage breast cancer: a decision model analysis.

British journal of cancer·2012
Same author

Clinically significant red blood cell antibodies in chronically transfused patients: a survey of Chinese thalassemia major patients and literature review.

Transfusion·2012
Same author

Epidemiological changes in hepatitis A in Korea: increasing age and its effect on clinical outcomes.

Epidemiology and infection·2012
Same author

Risk factor analysis of hepatitis C virus infection among Chinese blood donors in Hong Kong.

Transfusion medicine (Oxford, England)·2012
Same author

Successful treatment with anti-tumor necrosis factor alpha for reactive small-bowel amyloidosis.

Endoscopy·2011

This study introduces a novel two-stage method using multisignal minimum-cross-entropy spectral analysis (multisignal MCESA) to separate two simultaneous speech signals. The technique effectively refines spectral estimates, improving voice separation accuracy.

Area of Science:

  • Signal Processing
  • Speech Technology
  • Acoustics

Background:

  • Simultaneous speech on a single channel presents a significant challenge for signal separation.
  • Existing methods may struggle with accurately isolating individual talker spectra in cochannel environments.

Purpose of the Study:

  • To propose and evaluate a new two-stage approach for separating two simultaneously speaking talkers on a single channel.
  • To emphasize the implementation and effectiveness of the spectral tailoring procedure utilizing multisignal minimum-cross-entropy spectral analysis (multisignal MCESA).

Main Methods:

  • A two-stage system comprising a spectral separator followed by a spectral tailoring procedure.
  • The spectral tailoring procedure employs multisignal MCESA to refine initial spectral estimates of individual talkers.

Related Experiment Videos

  • Evaluation through listening tests and spectral distortion comparisons.

    • Main Results:

    • The spectral tailoring procedure, using multisignal MCESA, successfully adjusts initial spectral estimates.
    • Listening tests and spectral distortion analysis validate the usefulness of the proposed method.
    • Demonstrated improvement in separating speech signals from a single-channel composite source.

    Conclusions:

    • The proposed two-stage approach, particularly the spectral tailoring using multisignal MCESA, is effective for two-talker speech separation.
    • This method offers a viable solution for improving the clarity of individual speech signals in mixed-talker scenarios.
    • Further validation confirms the practical applicability of multisignal MCESA in advanced speech processing tasks.