Passer à la navigation principale Passer à la recherche Passer au contenu principal

Quasi-Newton method for maximum likelihood estimation of hidden Markov models

  • CNRS

Résultats de recherche: Le chapitre dans un livre, un rapport, une anthologie ou une collectionContribution à une conférenceRevue par des pairs

13 Citations (Scopus)

Résumé

Hidden Markov models (HMMs) are used in many signal processing applications including speech recognition, blind equalization of digital communications channels, etc. The most widely used method for maximum likelihood estimation of HMM parameters is the forward-backward (or Baum-Welch) algorithm which is an early example of application of the expectation-maximization (EM) principle. In this contribution, an alternative fast-converging approach for maximum likelihood estimation of HMM parameters is described. This new techniques is based on the use of general purpose quasi-Newton optimization methods as well as on an efficient purely recursive algorithm for computing the log-likelihood and its derivative.

langue originaleAnglais
titreProceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998
EditeurInstitute of Electrical and Electronics Engineers Inc.
Pages2265-2268
Nombre de pages4
ISBN (imprimé)0780344286, 9780780344280
Les DOIs
étatPublié - 1 janv. 1998
Evénement1998 23rd IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998 - Seattle, WA, États-Unis
Durée: 12 mai 199815 mai 1998

Série de publications

NomICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume4
ISSN (imprimé)1520-6149

Une conférence

Une conférence1998 23rd IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998
Pays/TerritoireÉtats-Unis
La villeSeattle, WA
période12/05/9815/05/98

Empreinte digitale

Examiner les sujets de recherche de « Quasi-Newton method for maximum likelihood estimation of hidden Markov models ». Ensemble, ils forment une empreinte digitale unique.

Contient cette citation