We propose a novel approach to the design of efficient representations of speech for various recognition tasks. Using a principled information theoretic framework - the Information Bottleneck method - which enables quantization that preserves relevant information, we demonstrate that significantly smaller representations of the signal can be obtained that still capture most of the relevant information about phonemes or speakers. The significant implications for building more efficient speech and speaker recognition systems are discussed.
Cite as: Hecht, R.M., Tishby, N. (2005) Extraction of relevant speech features using the information bottleneck method. Proc. Interspeech 2005, 353-356, doi: 10.21437/Interspeech.2005-182
@inproceedings{hecht05_interspeech,
author={Ron M. Hecht and Naftali Tishby},
title={{Extraction of relevant speech features using the information bottleneck method}},
year=2005,
booktitle={Proc. Interspeech 2005},
pages={353--356},
doi={10.21437/Interspeech.2005-182}
}