Detail publikace

Comparison of Different Kinds of Long-Time Spectra of Voice Estimated by Modified Linear Prediction to Distinguish Speakers

Originální název

Comparison of Different Kinds of Long-Time Spectra of Voice Estimated by Modified Linear Prediction to Distinguish Speakers

Anglický název

Comparison of Different Kinds of Long-Time Spectra of Voice Estimated by Modified Linear Prediction to Distinguish Speakers

Jazyk

en

Originální abstrakt

This paper deals with two kinds of long-time spectra of speech estimated by the linear prediction approach. The standard approach was modified in two ways to achieve the long-time effect - either autocorrelation coefficients (AC) or predictive coefficients (PC) were averaged over a period of 2 minutes. The spectra were computed using order of prediction from 6 to 22. To distinguish speakers, the most appropriate frequencies seem to be around 1010 Hz (AC averaged) or around 340 Hz (PC averaged).

Anglický abstrakt

This paper deals with two kinds of long-time spectra of speech estimated by the linear prediction approach. The standard approach was modified in two ways to achieve the long-time effect - either autocorrelation coefficients (AC) or predictive coefficients (PC) were averaged over a period of 2 minutes. The spectra were computed using order of prediction from 6 to 22. To distinguish speakers, the most appropriate frequencies seem to be around 1010 Hz (AC averaged) or around 340 Hz (PC averaged).

BibTex


@inproceedings{BUT159592,
  author="Milan {Sigmund}",
  title="Comparison of Different Kinds of Long-Time Spectra of Voice Estimated by Modified Linear Prediction to Distinguish Speakers",
  annote="This paper deals with two kinds of long-time spectra of speech estimated by the linear prediction approach. The standard approach was modified in two ways to achieve the long-time effect - either autocorrelation coefficients (AC) or predictive coefficients (PC) were averaged over a period of 2 minutes. The spectra were computed using order of prediction from 6 to 22. To distinguish speakers, the most appropriate frequencies seem to be around 1010 Hz (AC averaged) or around 340 Hz (PC averaged).",
  address="Romanian Academy, IEEE, EURASIP",
  booktitle="Proceedings of the 2019 International Conference on Speech Technology and Human-Computer Dialogue (SpeD) “SpeD 2019”",
  chapter="159592",
  doi="10.1109/SPED.2019.8906615",
  howpublished="online",
  institution="Romanian Academy, IEEE, EURASIP",
  year="2019",
  month="october",
  pages="1--6",
  publisher="Romanian Academy, IEEE, EURASIP",
  type="conference paper"
}