Title :
On the locality of the forward-backward algorithm [speech recognition]
Author :
Merialdo, Bernard
Author_Institution :
IBM France Sci. Center, Paris, France
fDate :
4/1/1993 12:00:00 AM
Abstract :
A theorem which shows that the local maximum found by the forward-backward algorithm in the case of discrete hidden Markov models is really local is presented. By this it is meant that this local maximum is restricted to lie in the same connected component of the set {x:P(x)⩾P(x0 )} as the initial point x0 (where P( x) is the polynomial being maximized). This theoretical result suggests that, in practice, the choice of the initial point is important for the quality of the maximum obtained by the algorithm
Keywords :
convergence of numerical methods; hidden Markov models; iterative methods; polynomials; speech recognition; Baum-Welch algorithm; HMM; discrete hidden Markov models; forward-backward algorithm; local maximum; locality; polynomial; speech recognition; Convergence; Hidden Markov models; Iterative algorithms; Loudspeakers; Polynomials; Speech recognition; Statistics; Training data;
Journal_Title :
Speech and Audio Processing, IEEE Transactions on
