• DocumentCode
    3597846
  • Title

    Single voice command recognition by finite element analysis

  • Author

    Todorov, T.D. ; Tsanev, G.S. ; Yordanov, S.Y.

  • Author_Institution
    Dept. of Math., Tech. Univ. of Gabrovo, Gabrovo, Bulgaria
  • fYear
    2014
  • Firstpage
    1
  • Lastpage
    4
  • Abstract
    The paper deals with voice recognition systems. A new finite element signal processing method is obtained. Each input signal is processed by a particular frequency. Most authors analyse the cepstral coefficients in their classifiers. The logarithmic energy vector is used for recognising the input signal in the present paper. A new finite element criteria for closeness between voice commands and preliminary stored patterns is introduced. For this purpose a new Quasi-Hermite interpolation operator is obtained. The new ideas are computer implemented. Real life tests supporting the considered theory have been carried out.
  • Keywords
    cepstral analysis; finite element analysis; interpolation; signal processing; speech recognition; vectors; cepstral coefficients; finite element signal processing method; logarithmic energy vector; particular frequency; quasi-Hermite interpolation operator; single voice command recognition; Interpolation; Cepstral coefficients; Mel filter bank; Quasi-Hermite interpolation operator; Score functional; Signal processing techniques;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Information Technology Based Higher Education and Training (ITHET), 2014
  • Type

    conf

  • DOI
    10.1109/ITHET.2014.7155712
  • Filename
    7155712