Chapter

Engineering of Intelligent Systems

Volume 2070 of the series Lecture Notes in Computer Science pp 502-512

Date:

Application of Feature Transformation and Learning Methods in Phoneme Classification

  • András KocsorAffiliated withResearch Group on Artificial Intelligence of the Hungarian Academy of Sciences and of the University of Szeged
  • , László TóthAffiliated withResearch Group on Artificial Intelligence of the Hungarian Academy of Sciences and of the University of Szeged
  • , László FelföldiAffiliated withResearch Group on Artificial Intelligence of the Hungarian Academy of Sciences and of the University of Szeged

* Final gross prices may vary according to local VAT.

Get Access

Abstract

This paper examines the applicability of some learning tech- niques to the classification of phonemes. The methods tested were arti- ficial neural nets (ANN), support vector machines (SVM) and Gaussian mixture modeling. We compare these methods with a traditional hid- den Markov phoneme model (HMM) working with the linear prediction- based cepstral coefficient features (LPCC). We also tried to combine the learners with feature transformation methods, like linear discriminant analysis (LDA), principal component analysis (PCA) and independent component analysis (ICA). We found that the discriminative learners can attain the efficiency of the HMM, and after LDA they can attain practically the same score on only 27 features. PCA and ICA proved ineffective, apparently because of the discrete cosine transform inherent in LPCC.