Maximum mutual information neural networks for hybrid connectionist-HMM speech recognition systems
Author:
- G. RigollGRG. Rigoll
- 1 January 1994
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Speech and Audio Processing
- Vol. 2 (1) , 175-184
Abstract
No abstract availableKeywords
This publication has 6 references indexed in Scilit:
- Using phoneme group specific LVQ-codebooks with HMMs1992
- Unsupervised information theory-based training algorithms for multilayer neural networks1992
- Neural networks for statistical inference: Generalizations with applications to speech recognition1991
- A new unsupervised learning algorithm for multilayer perceptrons based on information theory principles1991
- Speaker-independent large vocabulary word recognition using an LVQ/HMM hybrid algorithm1991
- Context-independent phonetic hidden Markov models for speaker-independent continuous speech recognitionIEEE Transactions on Acoustics, Speech, and Signal Processing, 1990