Automatic word recognition based on second-order hidden Markov models

We propose an extension of the Viterbi algorithm that makes second-order hidden Markov models computationally efficient. A comparative study between first-order (HMM1s) and second-order Markov models (HMM2s) is carried out. Experimental results show that HMM2s provide a better state occupancy modeli...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on speech and audio processing 1997-01, Vol.5 (1), p.22-25
Main Authors: Mari, J.-F., Haton, J.-P., Kriouile, A.
Format: Article
Language:English
Subjects:
Applied sciences
Computational complexity
Exact sciences and technology
Gaussian distribution
Hidden Markov models
Information, signal and communications theory
Random processes
Shape
Signal processing
Solid modeling
Speech processing
State estimation
Stochastic processes
Telecommunications and information theory
Viterbi algorithm
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We propose an extension of the Viterbi algorithm that makes second-order hidden Markov models computationally efficient. A comparative study between first-order (HMM1s) and second-order Markov models (HMM2s) is carried out. Experimental results show that HMM2s provide a better state occupancy modeling and, alone, have performances comparable with HMM1s plus postprocessing.
ISSN:1063-6676
1558-2353
DOI:10.1109/89.554265