Comparison of the order reducing (ORED) and fast incremental training (FIT) algorithms for training high order hidden Markov models

Simple item page
dc.contributor.authordu Preez J.A.
dc.contributor.authorWeber D.M.
dc.date.accessioned2011-05-15T15:57:30Z
dc.date.available2011-05-15T15:57:30Z
dc.date.issued1997
dc.description.abstractIn an accompanying paper we detailed the ORED mid FIT algorithms which are both applicable to the training of high order hidden Markov models (HMM). Due to the presence of local optima, the training algorithms are not guaranteed to converge to the same result. In this paper we use simulations as well as experiments on speech to investigate some differences between them. We show that the FIT algorithm requires a fraction of the computational requirements, while simultaneously providing better accuracy and generalization compared to the ORED approach. The experiments indicate that the FIT algorithm provides a practical approach to training high order HMMs in circumstances which might ordinarily be considered as unfeasible.
dc.description.versionConference Paper
dc.identifier.citationProceedings of the South African Symposium on Communications and Signal Processing, COMSIG
dc.identifier.urihttp://hdl.handle.net/10019.1/10442
dc.subjectComputer simulation
dc.subjectMarkov processes
dc.subjectMathematical models
dc.subjectFast incremental training (FIT) algorithms
dc.subjectHidden Markov models (HMM)
dc.subjectOrder reducing (ORED) algorithms
dc.subjectAlgorithms
dc.titleComparison of the order reducing (ORED) and fast incremental training (FIT) algorithms for training high order hidden Markov models
dc.typeConference Paper
Files
Collections
Stellenbosch University - Scopus Publications