Catalogo Articoli (Spogli Riviste)

OPAC HELP

Titolo:
A modified Baum-Welch algorithm for hidden Markov models with multiple observation spaces
Autore:
Baggenstoss, PM;
Indirizzi:
USN, Undersea Warfare Ctr, Newport, RI 02841 USA USN Newport RI USA 02841USN, Undersea Warfare Ctr, Newport, RI 02841 USA
Titolo Testata:
IEEE TRANSACTIONS ON SPEECH AND AUDIO PROCESSING
fascicolo: 4, volume: 9, anno: 2001,
pagine: 411 - 416
SICI:
1063-6676(200105)9:4<411:AMBAFH>2.0.ZU;2-0
Fonte:
ISI
Lingua:
ENG
Soggetto:
CHAINS;
Keywords:
Baum-Welch algorithm; class-specific; EM algorithm; expectation-maximization; Gaussian mixtures; hidden Markov model (HMM); parameter estimation; sufficient statistics;
Tipo documento:
Article
Natura:
Periodico
Settore Disciplinare:
Engineering, Computing & Technology
Citazioni:
11
Recensione:
Indirizzi per estratti:
Indirizzo: Baggenstoss, PM USN, Undersea Warfare Ctr, Newport, RI 02841 USA USN Newport RI USA 02841 rfare Ctr, Newport, RI 02841 USA
Citazione:
P.M. Baggenstoss, "A modified Baum-Welch algorithm for hidden Markov models with multiple observation spaces", IEEE SPEECH, 9(4), 2001, pp. 411-416

Abstract

In this paper, we derive an algorithm similar to the well-known Baum-Welchalgorithm for estimating the parameters of a hidden Markov model (HMM). The new algorithm allows the observation PDF of each state to be defined and estimated using: a different feature set, we show that estimating parameters in this manner is equivalent to maximizing the likelihood function for the standard parameterization of the HMM defined on the input data space, Theprocessor becomes optimal if the state-dependent feature sets are sufficient statistics to distinguish each state individually from a common state.

ASDD Area Sistemi Dipartimentali e Documentali, Università di Bologna, Catalogo delle riviste ed altri periodici
Documento generato il 30/03/20 alle ore 13:23:15