Speech/Non-Speech Classification in Hearing Aids Driven by Tailored Neural Networks
This chapter explores the feasibility of using some kind of tailored neural networks to automatically classify sounds into either speech or non-speech in hearing aids. These classes have been preliminary selected aiming at focusing on speech intelligibility and user's comfort. Hearing aids in the market have important constraints in terms of computational complexity and battery life, and thus a set of trade-offs have to be considered. Tailoring the neural network requires a balance consisting in reducing the computational demands (that is the number of neurons) without degrading the classification performance. Special emphasis will be placed on designing the size and complexity of the multilayer perceptron constructed by a growing method. The number of simple operations will be evaluated, to ensure that it is lower than the maximum sustained by the computational resources of the hearing aid.
KeywordsHearing Loss Feature Vector Digital Signal Processing Hide Neuron Audio Signal
Unable to display preview. Download preview PDF.
- 1.Büchler, M.: Algorithms for sound classification in hearing instruments. PhD thesis, Swiss Federal Institute of Technology, Zurich (2002)Google Scholar
- 3.Keidser, G.: Selecting different amplification for different listening conditions. Journal of the American Academy of Audiology 7 (1996) 92-104Google Scholar
- 4.Alexandre, E., Cuadra, L., Perez, L., Rosa-Zurera, M., Lopez-Ferreras, F.: Automatic sound classification for improving speech intelligibility in hearing aids using a layered structure. In: Lecture Notes in Computer Science. Springer Verlag, Berlin (2006)Google Scholar
- 5.Alexandre, E., Alvarez, L., Cuadra, L., Rosa, M., Lopez, F.: Automatic sound classification algorithm for hearing aids using a hierarchical taxonomy. In: Dobrucki, A., Petrovsky, A., Skarbek, W., eds.: New Trends in Audio and Video. Volume 1. Politechnika Bialostocka (2006)Google Scholar
- 6.Alexandre, E., Rosa, M., Lopez, F.: Application of Fisher Linear Discriminant Analysis to Speech/Music Classification. In: proceedings of EUROCON 2005 -The International Conference on “Computer as a tool”, Belgrade, Serbia and Montenegro, Volume: 2, ISBN: 1-4244-0049-XGoogle Scholar
- 7.Guaus, E., Batlle, E.: A non-linear rhythm-based style classification for broadcast speech-music discrimination. In: AES 116th Convention (2004)Google Scholar
- 9.Bishop, C.M.: Neural Networks for Pattern Recognition. Oxford University Press, Oxford (1995)Google Scholar
- 11.Foresee, F.D., Hagan, M.T.: Gauss-newton approximation to bayesian learning. Proceedings of the 1997 International Joint Conference on Neural Networks (1997) 1930-1935Google Scholar
- 12.Wanas, N., Auda, G., Kamel, M.S., Karray, F.: On the optimal number of hidden nodes in a neural network. In: IEEE Canadian Conference on Electrical and Computer Engineering (1998) 918-921Google Scholar
- 13.Mathia, K.: A variable structure learning algorithm for multilayer perceptrons. Intelligent Engineering Systems Through Artificial Neural Networks 14 (2004) 93-100Google Scholar
- 14.Pearce, B.Y.: Implementation of a variable structure neural network. Master’s Thesis, Auburn University (2001)Google Scholar