Detail publikace

Speaker Dependent Frequency Cepstrum Coefficients

Originální název

Speaker Dependent Frequency Cepstrum Coefficients

Anglický název

Speaker Dependent Frequency Cepstrum Coefficients

Jazyk

en

Originální abstrakt

This paper aims at speaker recognition based upon a novel set of features. Feature extraction is a crucial phase of the speaker recognition process and a proper feature set can influence it dramatically. Many well-known features are not suitable for the speaker recognition as those merge the specifics of the individual voices to make them universal. Therefore, we need features accentuating the individual differences of our voices to be able to recognise speakers reliably. This paper introduces Speaker Dependent Frequency Cepstrum Coefficients (SDFCC) intended for the speaker recognition purposes only. Experimental results prove increase of the reliability in comparison to the well-known features. According to the test results, the SDFCC are very useful and promising for the speaker recognition.

Anglický abstrakt

This paper aims at speaker recognition based upon a novel set of features. Feature extraction is a crucial phase of the speaker recognition process and a proper feature set can influence it dramatically. Many well-known features are not suitable for the speaker recognition as those merge the specifics of the individual voices to make them universal. Therefore, we need features accentuating the individual differences of our voices to be able to recognise speakers reliably. This paper introduces Speaker Dependent Frequency Cepstrum Coefficients (SDFCC) intended for the speaker recognition purposes only. Experimental results prove increase of the reliability in comparison to the well-known features. According to the test results, the SDFCC are very useful and promising for the speaker recognition.

BibTex


@inproceedings{BUT33733,
  author="Filip {Orság}",
  title="Speaker Dependent Frequency Cepstrum Coefficients",
  annote="This paper aims at speaker recognition based upon a novel set of features.
Feature extraction is a crucial phase of the speaker recognition process and
a proper feature set can influence it dramatically. Many well-known features are
not suitable for the speaker recognition as those merge the specifics of the
individual voices to make them universal. Therefore, we need features
accentuating the individual differences of our voices to be able to recognise
speakers reliably. This paper introduces Speaker Dependent Frequency Cepstrum
Coefficients (SDFCC) intended for the speaker recognition purposes only.
Experimental results prove increase of the reliability in comparison to the
well-known features. According to the test results, the SDFCC are very useful and
promising for the speaker recognition.",
  address="Springer Verlag",
  booktitle="Security Technology",
  chapter="33733",
  edition="CCIS 58",
  howpublished="print",
  institution="Springer Verlag",
  year="2009",
  month="december",
  pages="258--264",
  publisher="Springer Verlag",
  type="conference paper"
}