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

Training and search methods for speech recognition

F Jelinek1

  • 1IBM Thomas J. Watson Research Center, Yorktown Heights, NY 10598, USA.

Proceedings of the National Academy of Sciences of the United States of America
|October 24, 1995
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

Design of an endovascular morcellator for the surgical treatment of equine Cushing's disease.

The veterinary quarterly·2015
Same author

Vibrations in microtubules.

Journal of biological physics·2013
Same author

Mitochondrial metabolism - neglected link of cancer transformation and treatment.

Prague medical report·2012
Same author

Carcinoma of the trachea in a cat.

Journal of comparative pathology·2012
Same author

A maximum likelihood approach to continuous speech recognition.

IEEE transactions on pattern analysis and machine intelligence·2011
Same author

Adrenal glands of slaughtered bulls, heifers and cows: a histological study.

Anatomia, histologia, embryologia·2010
Same journal

Chemotactic self-organization captures the dynamics of mammalian hair follicle patterning.

Proceedings of the National Academy of Sciences of the United States of America·2026
Same journal

Tomographic imaging of superconducting order using particle-hole interference.

Proceedings of the National Academy of Sciences of the United States of America·2026
Same journal

Inhibitory potential of autologous neutralizing antibodies sets quantitative limits on the rebound-competent HIV-1 reservoir.

Proceedings of the National Academy of Sciences of the United States of America·2026
Same journal

Inferring epidemiological parameters under an infectious phylogeography model with visitor dynamics.

Proceedings of the National Academy of Sciences of the United States of America·2026
Same journal

Analytical modeling for suction cup designs for skin-interfaced wearable devices.

Proceedings of the National Academy of Sciences of the United States of America·2026
Same journal

Improving cell-free metabolism through direct integration of artificial respiratory chains.

Proceedings of the National Academy of Sciences of the United States of America·2026
See all related articles

This study details the Baum algorithm for Hidden Markov Models (HMMs) in speech recognition, focusing on parameter estimation from speech data. It also describes a practical search algorithm for identifying the most probable utterance from acoustic data.

Area of Science:

  • Computational Linguistics
  • Speech Processing
  • Machine Learning

Background:

  • Speech recognition relies on acoustic analysis, probability estimation, and utterance determination.
  • Hidden Markov Models (HMMs) are crucial for estimating the probability of acoustic index strings from speech data.

Purpose of the Study:

  • To describe the Baum algorithm for reestimating HMM parameters from speech data.
  • To present a practical search algorithm for determining the most probable utterance in speech recognition.

Main Methods:

  • Utilizes Hidden Markov Models (HMMs) for acoustic index string probability estimation.
  • Employs the Baum algorithm for HMM parameter reestimation.
  • Applies a Viterbi-based search algorithm for utterance recognition.

Related Experiment Videos

Main Results:

  • The Baum algorithm provides a method for obtaining HMM parameters from speech data.
  • The described search algorithm efficiently finds the most probable utterance, even with moderate vocabulary sizes.

Conclusions:

  • Accurate HMM parameter estimation via the Baum algorithm is vital for speech recognition.
  • Efficient search algorithms are necessary for practical speech recognition systems to identify the most likely utterance.