1st International ICST Workshop on Computational Forensics

Research Article

The Influence of Frame Length on Speaker Identification Performance

  • @INPROCEEDINGS{10.1109/IAS.2007.31,
        author={Donato  Impedovo and Mario Refice},
        title={The Influence of Frame Length on Speaker Identification Performance},
        proceedings={1st International ICST Workshop on Computational Forensics},
        publisher={IEEE},
        proceedings_a={IWCF},
        year={2007},
        month={9},
        keywords={Degradation  Feature extraction  Hidden Markov models  Impedance  Information security  Mel frequency cepstral coefficient  Signal processing  Speaker recognition  Speech  Testing},
        doi={10.1109/IAS.2007.31}
    }
    
  • Donato Impedovo
    Mario Refice
    Year: 2007
    The Influence of Frame Length on Speaker Identification Performance
    IWCF
    IEEE
    DOI: 10.1109/IAS.2007.31
Donato Impedovo1,*, Mario Refice1,*
  • 1: Dipartimento di Elettrotecnica ed Elettronica - Politecnico di Bari via Orabona 4, 70123 Bari, Italy
*Contact email: impedovo@deemail.poliba.it, refice@poliba.it

Abstract

Speaker recognition/identification is a challenge for the implementation of security applications. Unfortunately, degradation in performance is usually observed for high pitched speakers and also whenever average pitch varies significantly between enrolment and testing. In this paper, a study on the impact of the frame length used to extract features from speech signal on the performance of speaker identification is presented. Tests have been carried out on a text-dependent database. Results show that a combination of different frame sizes between the training and the recognition phases can cope with the degradation. A reduction between 40% and 65% in false rejections has been generally observed.