IMPROVING LEARNING ABILITY OF RECURRENT NEURAL NETWORKS - Experiments on Speech Signals of Patients with Laryngopathies

Jarosław Szkoła, Krzysztof Pancerz, Jan Warchoł

2011

Abstract

Recurrent neural networks can be used for pattern recognition in time series data due to their ability of memorizing some information from the past. The Elman networks are a classical representative of this kind of neural networks. In the paper, we show how to improve learning ability of the Elman network by modifying and combining it with another kind of a recurrent neural network, namely, with the Jordan network. The modified Elman-Jordan network manifests a faster and more exact achievement of the target pattern. Validation experiments were carried out on speech signals of patients with laryngopathies.

References

  1. Elman, J. (1990). Finding structure in time. Cognitive Science, 14:179-211.
  2. Jordan, M. (1986). Serial order: A parallel distributed processing approach. Technical Report 8604, University of California, San Diego, Institute for Cognitive Science.
  3. Mandic, D. and Chambers, J. (2001). Recurrent Neural Networks for Prediction: Learning Algorithms, Architectures and Stability. John Wiley & Sons, Ltd.
  4. Semmlow, J. (2009). Biosignal and Medical Image Processing. CRC Press.
  5. Szkola, J., Pancerz, K., and Warchol, J. (2010). Computerbased clinical decision support for laryngopathies using recurrent neural networks. In Proc. of the 10th International Conference on Intelligent Systems Design and Applications (ISDA'2010), Cairo, Egypt.
  6. Warchol, J. (2006). Speech Examination with Correct and Pathological Phonation Using the SVAN 912AE Analyser (in Polish). PhD thesis, Medical University of Lublin.
  7. Warchol, J., Szkola, J., and Pancerz, K. (2010). Towards computer diagnosis of laryngopathies based on speech spectrum analysis: A preliminary approach. In Fred, A., Filipe, J., and Gamboa, H., editors, Proc. of the Third International Conference on Bio-inspired Systems and Signal Processing (BIOSIGNALS'2010), pages 464-467, Valencia, Spain.
Download


Paper Citation


in Harvard Style

Szkoła J., Pancerz K. and Warchoł J. (2011). IMPROVING LEARNING ABILITY OF RECURRENT NEURAL NETWORKS - Experiments on Speech Signals of Patients with Laryngopathies . In Proceedings of the International Conference on Bio-inspired Systems and Signal Processing - Volume 1: BIOSIGNALS, (BIOSTEC 2011) ISBN 978-989-8425-35-5, pages 360-364. DOI: 10.5220/0003292603600364


in Bibtex Style

@conference{biosignals11,
author={Jarosław Szkoła and Krzysztof Pancerz and Jan Warchoł},
title={IMPROVING LEARNING ABILITY OF RECURRENT NEURAL NETWORKS - Experiments on Speech Signals of Patients with Laryngopathies},
booktitle={Proceedings of the International Conference on Bio-inspired Systems and Signal Processing - Volume 1: BIOSIGNALS, (BIOSTEC 2011)},
year={2011},
pages={360-364},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003292603600364},
isbn={978-989-8425-35-5},
}


in EndNote Style

TY - CONF
JO - Proceedings of the International Conference on Bio-inspired Systems and Signal Processing - Volume 1: BIOSIGNALS, (BIOSTEC 2011)
TI - IMPROVING LEARNING ABILITY OF RECURRENT NEURAL NETWORKS - Experiments on Speech Signals of Patients with Laryngopathies
SN - 978-989-8425-35-5
AU - Szkoła J.
AU - Pancerz K.
AU - Warchoł J.
PY - 2011
SP - 360
EP - 364
DO - 10.5220/0003292603600364