• DocumentCode
    2085769
  • Title

    Speaker recognition system and experiments with head / torso simulator and telephone transmission

  • Author

    Cetnarowicz, Damian ; Drgas, Szymon ; Dabrowski, Adam

  • Author_Institution
    Dept. of Control & Syst. Eng., Poznan Univ. of Technol., Poznań, Poland
  • fYear
    2010
  • fDate
    23-25 Sept. 2010
  • Firstpage
    99
  • Lastpage
    103
  • Abstract
    In this paper a system for speaker recognition and respective experiments based on telephone speech signal quality are presented and reported. First, the speech signals are transmitted using regular GSM and analog telephone system. The recorded signals are used as an input for the Gaussian mixture model based speaker recognition system. The results suggest that the parameters of MFCC extraction should be tailored to signal quality.
  • Keywords
    Gaussian processes; audio signal processing; cellular radio; speaker recognition; telephone sets; GSM; Gaussian mixture model; MFCC extraction; analog telephone system; head-torso simulator; speaker recognition; telephone signal transmission; telephone speech signal quality; Databases; GSM; Mel frequency cepstral coefficient; Speaker recognition; Speech; Speech recognition; Training;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Signal Processing Algorithms, Architectures, Arrangements, and Applications Conference Proceedings (SPA), 2010
  • Conference_Location
    Poznan
  • Print_ISBN
    978-1-4577-1485-6
  • Electronic_ISBN
    978-83-62065-07-3
  • Type

    conf

  • Filename
    5943722