Support Vector Machine Training for Improved Hidden Markov Modeling

We present a discriminative training algorithm, that uses support vector machines (SVMs), to improve the classification of discrete and continuous output probability hidden Markov models (HMMs). The algorithm uses a set of maximum-likelihood (ML) trained HMM models as a baseline system, and an SVM t...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on signal processing 2008-01, Vol.56 (1), p.172-188
Main Authors: Sloin, A., Burshtein, D.
Format: Article
Language:English
Subjects:
Algorithms
Applied sciences
Digits
Discriminative training
Error analysis
Exact sciences and technology
hidden Markov model (HMM)
Hidden Markov models
Information, signal and communications theory
Kernel
Mathematical models
Maximum likelihood estimation
Parameter estimation
Pattern recognition
Recognition
Signal and communications theory
Signal processing
Signal representation. Spectral analysis
Signal, noise
Speech analysis
Speech processing
Speech recognition
support vector machine (SVM)
Support vector machine classification
Support vector machines
Tasks
Telecommunications and information theory
Training
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 present a discriminative training algorithm, that uses support vector machines (SVMs), to improve the classification of discrete and continuous output probability hidden Markov models (HMMs). The algorithm uses a set of maximum-likelihood (ML) trained HMM models as a baseline system, and an SVM training scheme to rescore the results of the baseline HMMs. It turns out that the rescoring model can be represented as an unnormalized HMM. We describe two algorithms for training the unnormalized HMM models for both the discrete and continuous cases. One of the algorithms results in a single set of unnormalized HMMs that can be used in the standard recognition procedure (the Viterbi recognizer), as if they were plain HMMs. We use a toy problem and an isolated noisy digit recognition task to compare our new method to standard ML training. Our experiments show that SVM rescoring of hidden Markov models typically reduces the error rate significantly compared to standard ML training.
ISSN:1053-587X
1941-0476
DOI:10.1109/TSP.2007.906741