Optimal On-Line Bayesian Model Selection for Speaker Adaptation

Document Type

Conference Proceeding

Publication Date


Find in a Library

Catalog Record


In this paper, we show how to accommodate a Bayesian variant of Rissanen's MDL into on-line Bayesian adaptation to control both model structural complexity and parameterization complexity to best fit an available amount of adaptation data, the goal being minimization of resulting recognition error. An efficient bottom-up dynamic programming based pruning algorithm is developed for selecting models using the MDL principle. Speaker adaptation experiments using a 26-letter English alphabet vocabulary were conducted and the proposed Bayesian variant MDL method is shown to provide an optimal trade-off between recognition accuracy and complexity of model structure and parameterization over a full range of adaptation data size. It in general is capable of automatically selecting a set of model parameters that leads to best recognition performance for a given amount of adaptation data.


Presented at the Sixth International Conference on Spoken Language Processing (ICSLP), Beijing, China, October 16-20, 2000.

Catalog Record