|
KMID : 0917520000070040011
|
|
Journal of Speech Sciences
2000 Volume.7 No. 4 p.11 ~ p.26
|
|
|
Semi-Continuous Hidden Markov Model with the MIN Module
|
|
김대극/Kim, Dae Keuk
이정주/정호균/이상희/Lee, Jeong Ju/Jeong, Ho Kyoun/Lee, Sang Hee
|
|
|
Abstract
|
|
|
In this paper, we propose the HMM with the MIN module. Because initial and re-estimated variane vectors are important elements for performance in HMM recognition systems, we propose a method which compensates for the mismatched statistical feature of training and test data. The MIN module function is a differentiable function similar to the sigmoid function. Unlike a continuous density function, it does not include variance vectors of the data set. The proposed hybrid HMM/MIN module is a unified network in which the observation probability in the HMM is replaced by MIN module neural network. The parameters in the unified network are re-estimated by the gradient descent method for the Maximum Likelihood (ML) criterion. In estimating parameters, the variance vector is not estimated beacuse there is no variance element in the MIN module function.
The experiment was peroformed to compare the performance of the proposed HMM and the conventional HMM. The experiment measured an isolated number for speaker independent recognition.
Keywords : Hidden Markov Model, MIN module, Maximum Likelihood
|
|
|
KEYWORD
|
|
|
|
|
|
FullTexts / Linksout information
|
|
|
|
|
|
Listed journal information
|
|
|
|
|