Learning the structure of HMM's through grammatical inference techniques
- 4 December 2002
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 717-720 vol.2
- https://doi.org/10.1109/icassp.1990.115882
Abstract
A technique is described in which all the components of a hidden Markov model are learnt from training speech data. The structure or topology of the model (i.e. the number of states and the actual transitions) is obtained by means of an error-correcting grammatical inference algorithm (ECGI). This structure is then reduced by using an appropriate state pruning criterion. The statistical parameters that are associated with the obtained topology are estimated from the same training data by means of the standard Baum-Welch algorithm. Experimental results showing the applicability of this technique to speech recognition are presented.<>Keywords
This publication has 7 references indexed in Scilit:
- Learning accurate finite-state structural models of words through the ECGI algorithmPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- Some relations among stochastic finite state networks used in automatic speech recognitionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1990
- On a model-robust training method for speech recognitionIEEE Transactions on Acoustics, Speech, and Signal Processing, 1988
- Recent Advances in Speech Understanding and Dialog SystemsPublished by Springer Nature ,1988
- Modelling (Sub)String Length Based Constraints through a Grammatical Inference MethodPublished by Springer Nature ,1987
- Experiments in dynamic programming inference of Markov networks with strings representing speech dataPattern Recognition, 1986
- Grammatical Inference: Introduction and Survey - Part IIEEE Transactions on Systems, Man, and Cybernetics, 1975