Continuous speech recognition using PLP analysis with multilayer perceptrons

The authors investigate the use of continuous features derived by perceptual linear predictive (PLP) analysis, examine the effect of adding temporal features, and compare it to the previously studied use of multiframe input. Comparisons of the MLP (multilayer perceptron) and conventional Gaussian cl...

Full description

Saved in:
Bibliographic Details
Main Authors: Morgan, N., Hermansky, H., Bourlard, H., Kohn, P., Wooters, C.
Format: Conference Proceeding
Language:English
Subjects:
Cepstral analysis
Computer science
Hidden Markov models
Information analysis
Multilayer perceptrons
Resource management
Spatial databases
Speech analysis
Speech recognition
Training data
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The authors investigate the use of continuous features derived by perceptual linear predictive (PLP) analysis, examine the effect of adding temporal features, and compare it to the previously studied use of multiframe input. Comparisons of the MLP (multilayer perceptron) and conventional Gaussian classifiers are also reported. The speaker-dependent portion of the Resource Management database was used for this test. Additionally, some experiments were performed with a perplexity-2200 speaker-independent recognition task on a subset of the TIMIT database. In each case, the PLP features were used as input to the networks. The experiments show the advantage of continuous PLP features and their first and second temporal derivatives.< >
ISSN:1520-6149
2379-190X
DOI:10.1109/ICASSP.1991.150275