System low speed encode the speech signal at a variable duration segment analysis

Authors

  • Андрей Алексеевич Афанасьев Academy FGS of Russia (Orel)
  • Роман Сергеевич Власов Academy FGS of Russia (Orel)

DOI:

https://doi.org/10.17308/sait.2019.3/1317

Keywords:

речевой сигнал, сегмент анализа, линейные спектральные частоты, шаг квантования, расширение сегмента

Abstract

In article, materials on research of a speech signal such properties, as uniformity (similitude) of the speech signal instant values distribution are stated. This property is used in a way of the speech processing segments allocation on the basis of the consecutive statistical analysis. The parameters of a linear prediction counted on each step of expansion undergo insignificant (within a quantization step) changes in relation to an initial segment. This circumstance allows to use results of the analysis of an initial site for restoration of the site expanded in the above-stated way that allows to lower essentially both computing complexity, and speed of a speech signal coding.

Author Biographies

  • Андрей Алексеевич Афанасьев, Academy FGS of Russia (Orel)

    Candidate of Technical Sciences, Assistant professor, Academy FGS of Russia, city Orel

  • Роман Сергеевич Власов, Academy FGS of Russia (Orel)

    Employee, Academy FGS of Russia, city Orel

References

Downloads

Published

2019-06-28

Issue

Section

Computer Linguistics and Natural Language Processing

How to Cite

System low speed encode the speech signal at a variable duration segment analysis. (2019). Proceedings of Voronezh State University. Series: Systems Analysis and Information Technologies, 3, 166-176. https://doi.org/10.17308/sait.2019.3/1317