Abstract
The paper present an application of one method for obtaining derivatives used in training a recurrent neural network that combine the technique of backpropagation through time and dual heuristic programming. The recurrent network was used in contextual causal phoneme recognition for complete (phoneme segmented) training set labeling.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Bridle, J.S.: ALPHA-NETS: A Recurrent Neural Network Architecture with a Hidden Markov Model Interpretation. Speech Communication, 9 (1992) 83–92.
Bridle, J.S., Dodd, S.: An Alphanet Approach to Optimizing Input Transformations for Continuos Speech Recognition. In: Proceedings ICASSP’ 90 (1990) 277–280.
Cichocki, A., Unbehauen, R.: Neural Networks for Optimization and Signal Processing. J.Wiley & Sons, New York (1993).
Filasová, A., Krokavec, D.: Heuristic Dynamic Programming for LQR Controllers and Kalman Estimators. In: Proceedings of the Conference on Artificial Intelligence in Industry AIII’ 98. High Tatras, Slovak Republic (1998) 110–119.
Herz, J., Krogh, A., Palmer, R.G.: Introduction to the Theory of Neural Computation. Addison-Wesley, Redwood City (1991).
Krokavec, D.: Neural Networks for Phoneme Probability Estimation. In: Proceedings of the Conference on Artificial Intelligence in Industry AIII’ 98. High Tatras, Slovak Republic (1998) 182–191.
Krokavec, D., Saxa, J.: Recursive Neural Network for Phoneme Recognition. In: Proceedings of the 3rd Conference on Digital Signal Processing DSP’97. Herlany, Slovak Republic (1997) 5–8.
Lee, K.F., Hon, H.W.: Speaker-Independent Phone Recognition Using Hidden Markov Models. IEEE Transactions on Acoustics, Speech and Signal Processing, 37 (1989) 1641–1648.
Matoušek, V.: Phonetic Segmentation Method for the Continuous Czech SpeechRecognition. In: Proceedings EUROSPEECH’ 93 (1993) 713–717.
Richard, M.D., Lippmann, R.P.: Neural Network Classifiers Estimate Bayesian a Posteriori Probabilities. Neural Computation, 3 (1991) 461–483.
Riis, S.K., Krogh, A.: Hidden Neural Networks: A Framework for HMM/NN Hybrids. In: Proceedings of ICASSP’ 97 (1997) 3233–3236.
Robinson, T.: An Application of Recurrent Nets to Phone Probability Estimation. IEEE Transactions on Neural Networks, 5 (1994). 298–305
Werbos, P.J.: Backpropagation through Time: What It Does and How to Do It. Proceedings of the IEEE, 78 (1990) 1550–1560.
Werbos, P.J.: Consistency of HDP Applied to a Simple Reinforcement Learning Problem. Neural Networks, 3 (1990) 179–189.
Young, S.J.: Competitive Training in Hidden Markov Models. In: Proceedings ICASSP’ 90 (1990) 681–684.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1999 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Krokavec, D., Filasová, A. (1999). State-Space Model Based Labeling of Speech Signals. In: Matousek, V., Mautner, P., Ocelíková, J., Sojka, P. (eds) Text, Speech and Dialogue. TSD 1999. Lecture Notes in Computer Science(), vol 1692. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-48239-3_47
Download citation
DOI: https://doi.org/10.1007/3-540-48239-3_47
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-66494-9
Online ISBN: 978-3-540-48239-0
eBook Packages: Springer Book Archive