Cover
Vol. 10 No. 1 (2010)

Published: June 30, 2010

Pages: 46-53

Original Article

Hierarchical Phoneme Recognition Using Node-wise Relevance-Optimized Features

Ali E. Hameed 1 lntessar T. Hwaidy 1
1 Department of Computer Engineering, College of Engineering, University cf Basrah
History
  • Received: March 30, 2010
  • Available Online: June 30, 2010
DOI

https://doi.org/10.33971/bjes.10.1.5

Abstract

Tn this paper, a hierarchical phoneme recognition system is proposed. The hierarchical approach is applied here to recursively partition the recognition problem into smaller and smaller sub-problems those are independently handled at the di~tinct nodes of the hierarchy. The nodes are individually set to chara~erize different properties of the input phoneme, or more precisely to make separate d~isioos on its pertinence to the different reference subgroups of phonemes. The full characteri:zation of the input phoneme is achieved by traversing some root-to-leaf path through the hierarchy. The relationships between the different features of phonemes and their pertinence to the different reference subgroups. are to be objectively characteriz.ed ttnd optimized here. This involves specifying the decisive subset of features for each pertinence decision and neglecting the remaining features those are irrelevam to (or probably have negative effect on) that decision, at each node of the hierarchy. The optimization applied through the feature election process here, is not aimed at reducing the amount of feamres to be used in the recognition process, for the purpose of decreasing the time-complexity of the systcn1, but, is interested in enhancing the decision making accuracy of the system by avoiding the misleading features.

Keywords

References

  1. S. R. Safavian and D. Landgrebe, "A Survey of Decision Tree Classifier Methodology", IEEE Transactions on Syst~ms. Man, & Cybernetics, vol. 21, no. 3, pp. 660-674, May 1991.
  2. S. R. Waterhouse, "Divide and Conquer Pattern Recognition using Mixtures of Experts", Ph.D. Dissertation, Jesus College, University of Cambridge, and Cambridge Univer.tity Engineering Dep(1rtrnent, feb 1997.
  3. A A. Ali and I. T. Hwaidy, "Hierarchical Arabic Phoneme Recognition using MFCC Analysis", Iraqi Journal for Electrical and Electronic Engineering, vol. 3, no. I, pp. 97-106, 2007.
  4. I. T. Ilwaidy, "Hierarchical Arabic Phoneme Recognition using Neural Networks", M.Sc. Tbesis, Computer Engineering Department. Engineering College, University of Basruh, March 2006.
  5. L. Rabinar and R. W. Schafer, ''Fundamentals of Speech Recognition", Prentice Hall, 1993.
  6. J. Y. Siva Rama Krishna Rao, "Recognition of Consonant-Vowel (CV) Utterances using Modular Neural Network Models", M.Sc. ]hesis, Department of Computer Science and Engineering, Indian inslirute of Technology, Madr(,f$, May 2000.
  7. M. Riedmiller and H. Braun, "A Direct Adaptive Method for Faster Backpropagation Learning: The RPROP Algorithm", Proceedings of the IEEE International Conference on Neural Network, San Francisco, pp. 586-591, 1993.
  8. M. Riedmiller, "Advanced Supervised Leaming in Multi-layer Perceptrons -from Backpropagation to Adaptive Learning Algorithms", International Journal of Computer Standard and Interface, vol. 16, pp. 265-278. 1994.
  9. P. Melin, J. Urias, D. Solano, M. SolO, M. Lope7. and 0. Caslillo, "Voice Recognition with Neuml Networks, Typc-2 fuzzy Logic and Genetic Algorithms", Engineering Letters, 13:2, EL_13_2_9, Advance on/i11e publication, August 2006.
  10. A. A. Jasim, "Hybrid Approaches for Arabic Phoneme Recognition", M.Sc. Thesis, Computer Engineering Department, Engineering College, University of nasrah, 2004.