Soft state-tying for HMM-based speech recognition

Christoph Neukirchen, Daniel Willett, Gerhard Rigoll
1998 5th International Conference on Spoken Language Processing (ICSLP 1998)   unpublished
This paper introduces a method for regularization of HMM systems that avoids parameter overfitting caused by insufficient training data. Regularization is done by augmenting the EM training method by a penalty term that favors simple and smooth HMM systems. The penalty term is constructed as a mixture model of negative exponential distributions that is assumed to generate the state dependent emission probabilities of the HMMs. This new method is the successful transfer of a well known
more » ... tion approach in neural networks to the HMM domain and can be interpreted as a generalization of traditional state-tying for HMM systems. The effect of regularization is demonstrated for continuous speech recognition tasks by improving overfitted triphone models and by speaker adaptation with limited training data.
doi:10.21437/icslp.1998-195 fatcat:job6dtjyqnbcxmmqfmdee5qaiq