Integrating Recent MLP Feature Extraction Techniques into TRAP Architecture

Abstract

This paper is focused on the incorporation of recent techniques for multi-layer perceptron (MLP) based feature extraction in Temporal Pattern (TRAP) and Hidden Activation TRAP (HATS) feature extraction scheme. The TRAP scheme has been origin of various MLP-based features some of which are now indivisible part of state-of-the-art LVCSR systems. The modifications which brought most improvement – sub-phoneme targets and Bottle-Neck technique – are introduced into original TRAP scheme. Introduction of sub-phoneme targets uncovered the hidden danger of having too many classes in TRAP/HATS scheme. On the other hand, Bottle-Neck technique improved the TRAP/HATS scheme so its competitive with other approaches.

Extracted Key Phrases

6 Figures and Tables

Cite this paper

@inproceedings{Grzl2011IntegratingRM, title={Integrating Recent MLP Feature Extraction Techniques into TRAP Architecture}, author={Frantisek Gr{\'e}zl and Martin Karafi{\'a}t}, booktitle={INTERSPEECH}, year={2011} }