Please use this identifier to cite or link to this item:
doi:10.22028/D291-25196
Title: | Neural networks for nonlinear discriminant analysis in continuous speech recognition |
Author(s): | Reichl, W. Harengel, S. Wolfertstetter, F. Ruske, G. |
Language: | English |
Year of Publication: | 1996 |
OPUS Source: | Saarbrücken, 1996 |
SWD key words: | Künstliche Intelligenz |
DDC notations: | 004 Computer science, internet |
Publikation type: | Report |
Abstract: | In this paper neural networks for Nonlinear Discriminant Analysis in continuous speech recognition are presented. Multilayer Perceptrons are used to estimate a-posteriori probabilities for Hidden-Markov Model states, which are the optimal discriminant features for the separation of the HMM states. The a-posteriori probabilities are transformed by a principal component analysis to calculate the new features for semicontinuous HMMs, which are trained by the known Maximum-Likelihood training. The nonlinear discriminant transformation is used in speaker-independent phoneme recognition experiments and compared to the standard Linear Discriminant Analysis technique. |
Link to this record: | urn:nbn:de:bsz:291-scidok-41936 hdl:20.500.11880/25252 http://dx.doi.org/10.22028/D291-25196 |
Series name: | Vm-Report / Verbmobil, Verbundvorhaben, [Deutsches Forschungszentrum für Künstliche Intelligenz] |
Series volume: | 111 |
Date of registration: | 6-Sep-2011 |
Faculty: | SE - Sonstige Einrichtungen |
Department: | SE - DFKI Deutsches Forschungszentrum für Künstliche Intelligenz |
Collections: | SciDok - Der Wissenschaftsserver der Universität des Saarlandes |
Files for this record:
File | Description | Size | Format | |
---|---|---|---|---|
report_111_96.pdf | 279,2 kB | Adobe PDF | View/Open |
Items in SciDok are protected by copyright, with all rights reserved, unless otherwise indicated.