• Computer Science
  • Published 1994

NON-LINEAR INPUT TRANSFORMATIONS FOR DISCRIMINATIVE HMMS

@inproceedings{Tore1994NONLINEARIT,
  title={NON-LINEAR INPUT TRANSFORMATIONS FOR DISCRIMINATIVE HMMS},
  author={Finn Tore},
  year={1994}
}
This paper deals wir h speaker-independent continuous speech recognition. Our approach is based on continuous density hidden Markov models with a non-linpar input featmre transformation performed by a multilayer perceptron. We discuss various optimisation criteria and provide results on a TIMIT phoneme recognition task, using single frame MMI embedded in Viterbi training, and a global MMI criterion. As expected, global MMJ is found superior to the frame-based criterion for continuous… CONTINUE READING

Citations

Publications citing this paper.
SHOWING 1-10 OF 10 CITATIONS

Hidden Markov models and neural networks for speech recognition

VIEW 7 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

Hidden neural networks: a framework for HMM/NN hybrids

VIEW 4 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

Joint Estimation of Parameters in Hidden Neural Networks

VIEW 5 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Hidden Neural Networks

VIEW 2 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Hidden neural networks: application to speech recognition

  • Søren Kamaric Riis
  • Computer Science
  • Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '98 (Cat. No.98CH36181)
  • 1998
VIEW 3 EXCERPTS
CITES METHODS & RESULTS
HIGHLY INFLUENCED

Automatic signature verification system

VIEW 1 EXCERPT
CITES BACKGROUND

A survey of hybrid ANN/HMM models for automatic speech recognition

VIEW 1 EXCERPT
CITES METHODS

Refining hidden Markov models with recurrent neural networks

  • T. Wessels, Christian W. Omlin
  • Computer Science
  • Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Networks. IJCNN 2000. Neural Computing: New Challenges and Perspectives for the New Millennium
  • 2000

Characterization of speakers for improved automatic speech recognition

VIEW 2 EXCERPTS
CITES BACKGROUND & METHODS