Using mutual information to design feature combinations


Combination of different feature streams is a well-established method for improving speech recognition performance. This empirical success, however, poses theoretical problems when trying to design combination systems: is it possible to predict which feature streams will combine most advantageously, and which of the many possible combination strategies will… (More)


4 Figures and Tables


Citations per Year

64 Citations

Semantic Scholar estimates that this publication has 64 citations based on the available data.

See our FAQ for additional information.

Cite this paper

@inproceedings{Ellis2000UsingMI, title={Using mutual information to design feature combinations}, author={Daniel P. W. Ellis and Jeff A. Bilmes}, booktitle={INTERSPEECH}, year={2000} }