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

Continuous speech recognition by connectionist statistical methods.

H Bourlard1, N Morgan

  • 1Int. Comput. Sci. Inst., Berkeley, CA.

IEEE Transactions on Neural Networks
|January 1, 1993
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

Research note: Expert opinions of feather sucking and licking behavior in meat chicken breeder birds.

Poultry science·2024
Same author

Elastohydrodynamic lubricant flow with nanoparticle tracking.

RSC advances·2022
Same author

Non-starch polysaccharide degradation in the gastrointestinal tract of broiler chickens fed commercial-type diets supplemented with either a single dose of xylanase, a double dose of xylanase, or a cocktail of non-starch polysaccharide-degrading enzymes.

Poultry science·2022
Same author

Audit of vaccination status of health-care workers who tested positive for SARS-CoV-2.

Journal of clinical virology plus·2022
Same author

Zinc hydroxychloride supplementation improves tibia bone development and intestinal health of broiler chickens.

Poultry science·2021
Same author

Dietary soluble non-starch polysaccharide level and composition influences grower and finisher phase performance, excreta moisture content and total tract nutrient digestibility in broilers.

British poultry science·2021
Same journal

Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

IEEE transactions on neural networks·2013
Same journal

Guest editorial: special section on white box nonlinear prediction models.

IEEE transactions on neural networks·2011
Same journal

Data-based fault-tolerant control of high-speed trains with traction/braking notch nonlinearities and actuator failures.

IEEE transactions on neural networks·2011
Same journal

Guest editorial: special section on data-based control, modeling, and optimization.

IEEE transactions on neural networks·2011
Same journal

Neural network-based multiple robot simultaneous localization and mapping.

IEEE transactions on neural networks·2011
Same journal

Data-driven model-free adaptive control for a class of MIMO nonlinear discrete-time systems.

IEEE transactions on neural networks·2011
See all related articles

Multilayer perceptrons (MLP) effectively estimate probability density functions for continuous speech recognition. This hybrid hidden Markov model (HMM)/MLP approach offers improvements for context-dependent speech recognition systems.

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Speech Recognition

Background:

  • Early speech recognition relied on context-independent hidden Markov models (HMM).
  • Multilayer perceptrons (MLP) show promise for estimating high-dimensional probability density functions.
  • Extending HMMs to context-dependent models is crucial for improved speech recognition.

Purpose of the Study:

  • To review the principles of a hybrid HMM/MLP approach for continuous speech recognition.
  • To describe recent improvements to the hybrid HMM/MLP system.
  • To evaluate the performance of the enhanced system on a standard speech database.

Main Methods:

  • Utilizing multilayer perceptrons (MLP) for probability density function estimation.
  • Developing context-dependent hidden Markov models (HMM).

Related Experiment Videos

  • Implementing system modifications analogous to conventional HMM systems, balancing computational complexity with memory requirements.

    • Main Results:

    • The hybrid HMM/MLP approach demonstrates effectiveness in continuous speech recognition.
    • Improvements enhance system performance, particularly for context-dependent models.
    • Results are validated on the Resource Management speech database.

    Conclusions:

    • The hybrid HMM/MLP method is a viable and effective approach for advanced speech recognition.
    • System enhancements offer trade-offs between computational load and memory efficiency.
    • The approach shows significant potential for future speech recognition technologies.