We present a novel method of dimension reduction and feature selection that makes use of category-dependent regions in highdimensional data. Our method is inspired by phoneme-dependent, noise-robust low-variance regions observed in the cortical response, and introduces the notion of category-dependence in a two-step dimension reduction process that draws on the fundamental principles of Fisher Linear Discriminant Analysis. As a method of applying these features in an actual pattern classification task, we construct a system of multiple speech recognizers that are combined by a Bayesian decision rule under some simplifying assumptions. The results show a significant increase in recognition rate for low signal-to-noise ratios compared with previous methods, providing motivation for further study on hierarchical, category-dependent recognition and detection.
Cite as: Jeon, W., Juang, B.-H. (2005) A category-dependent feature selection method for speech signals. Proc. Interspeech 2005, 365-368, doi: 10.21437/Interspeech.2005-185
@inproceedings{jeon05_interspeech,
author={Woojay Jeon and Biing-Hwang Juang},
title={{A category-dependent feature selection method for speech signals}},
year=2005,
booktitle={Proc. Interspeech 2005},
pages={365--368},
doi={10.21437/Interspeech.2005-185}
}