Decoder selection based on cross-entropies

The authors generalize the maximum likelihood and related optimization criteria for training and decoding with a speech recognizer. The generalizations are constructed by considering weighted linear combinations of the logarithms of the likelihoods of words, of acoustics, and of (word, acoustic) pai...

Full description

Saved in:
Bibliographic Details
Published inICASSP-88., International Conference on Acoustics, Speech, and Signal Processing pp. 20 - 23 vol.1
Main Authors Gopalakrishnan, P.S., Kanevsky, D., Nadas, A., Nahamoo, D., Picheny, M.A.
Format Conference Proceeding
LanguageEnglish
Published IEEE 1988
Subjects
Acoustics
Computer science
Error analysis
Maximum likelihood decoding
Maximum likelihood estimation
Mutual information
Probability distribution
Random variables
Speech recognition
Online AccessGet full text

Cover

More Information
Summary:The authors generalize the maximum likelihood and related optimization criteria for training and decoding with a speech recognizer. The generalizations are constructed by considering weighted linear combinations of the logarithms of the likelihoods of words, of acoustics, and of (word, acoustic) pairs. The utility of various patterns of weights are examined.< >
ISSN:1520-6149
2379-190X
DOI:10.1109/ICASSP.1988.196499