Research Article Open Access

Smooth Formant Peak Via Discrete Tchebichef Transform

Nur Azman Abu1, Ferda Ernawan2 and Fadzilah Salim1
  • 1 Universiti Teknikal Malaysia Melaka, Malaysia
  • 2 Universitas Dian Nuswantoro, Indonesia
Journal of Computer Science
Volume 11 No. 2, 2015, 351-360

DOI: https://doi.org/10.3844/jcssp.2015.351.360

Published On: 10 September 2014

How to Cite: Abu, N. A., Ernawan, F. & Salim, F. (2015). Smooth Formant Peak Via Discrete Tchebichef Transform. Journal of Computer Science, 11(2), 351-360. https://doi.org/10.3844/jcssp.2015.351.360

Abstract

With the growth in computing power, speech recognition carries a strong potential in the near future. It has even become increasingly popular with the development of mobile devices. Presumably, mobile devices have limited computational power, memory size and battery life. In general, speech recognition operation requires heavy computation due to large samples per window used. Fast Fourier Transfom (FFT) is the most popular transform to search for formant frequencies in speech recognition. In addition, FFT operates in complex fields with imaginary numbers. This paper proposes an approach based on Discrete Tchebichef Transform (DTT) as a possible alternative to FFT in searching for the formant frequencies. The experimental outputs in terms of the frequency formants using FFT and DTT have been compared. Interestingly, the experimental results show that both have produced relatively identical formant shape output in terms of basic vowels and consonants recognition. DTT has the same capability to recognize speech formants F1, F2, F3 on real domains.

Download

Keywords

  • Formant Estimation
  • Discrete Tchebichef Transform
  • Spectrum Analysis
  • Fast Fourier Transform
  • Orthogonal Transform Function