Stochastic modeling of spectral adjustment for high quality pitch modification

Alexander Kain, Yannis Stylianou

Research output: Chapter in Book/Report/Conference proceedingConference contribution

20 Citations (Scopus)

Abstract

We present a new algorithm for adjusting the magnitude spectrum when the fundamental frequency (F0) of a speech signal is altered. The algorithm exploits the correlation between F0 and the magnitude spectrum of speech as represented by line spectral frequencies (LSFs). This correlation is class-dependent, and thus a broad classification of the input is achieved by a Gaussian mixture model (GMM). The within-class dependencies of LSFs on F0 values are captured by constructing their joint probability densities using a series of GMMs, one for each speech class. The proposed system is used for post-processing the pitch modified signal. Perceptual tests showed that the addition of this post-processing system improves the naturalness of the pitch modified signal for large pitch modification factors.

Original languageEnglish (US)
Title of host publicationICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
PublisherIEEE
Pages949-952
Number of pages4
Volume2
StatePublished - 2000
Event2000 IEEE Interntional Conference on Acoustics, Speech, and Signal Processing - Istanbul, Turkey
Duration: Jun 5 2000Jun 9 2000

Other

Other2000 IEEE Interntional Conference on Acoustics, Speech, and Signal Processing
CityIstanbul, Turkey
Period6/5/006/9/00

Fingerprint

adjusting
line spectra
Processing

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering
  • Acoustics and Ultrasonics

Cite this

Kain, A., & Stylianou, Y. (2000). Stochastic modeling of spectral adjustment for high quality pitch modification. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings (Vol. 2, pp. 949-952). IEEE.

Stochastic modeling of spectral adjustment for high quality pitch modification. / Kain, Alexander; Stylianou, Yannis.

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Vol. 2 IEEE, 2000. p. 949-952.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kain, A & Stylianou, Y 2000, Stochastic modeling of spectral adjustment for high quality pitch modification. in ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. vol. 2, IEEE, pp. 949-952, 2000 IEEE Interntional Conference on Acoustics, Speech, and Signal Processing, Istanbul, Turkey, 6/5/00.
Kain A, Stylianou Y. Stochastic modeling of spectral adjustment for high quality pitch modification. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Vol. 2. IEEE. 2000. p. 949-952
Kain, Alexander ; Stylianou, Yannis. / Stochastic modeling of spectral adjustment for high quality pitch modification. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Vol. 2 IEEE, 2000. pp. 949-952
@inproceedings{7c9fca3b339747daa260bef3b56bac21,
title = "Stochastic modeling of spectral adjustment for high quality pitch modification",
abstract = "We present a new algorithm for adjusting the magnitude spectrum when the fundamental frequency (F0) of a speech signal is altered. The algorithm exploits the correlation between F0 and the magnitude spectrum of speech as represented by line spectral frequencies (LSFs). This correlation is class-dependent, and thus a broad classification of the input is achieved by a Gaussian mixture model (GMM). The within-class dependencies of LSFs on F0 values are captured by constructing their joint probability densities using a series of GMMs, one for each speech class. The proposed system is used for post-processing the pitch modified signal. Perceptual tests showed that the addition of this post-processing system improves the naturalness of the pitch modified signal for large pitch modification factors.",
author = "Alexander Kain and Yannis Stylianou",
year = "2000",
language = "English (US)",
volume = "2",
pages = "949--952",
booktitle = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",
publisher = "IEEE",

}

TY - GEN

T1 - Stochastic modeling of spectral adjustment for high quality pitch modification

AU - Kain, Alexander

AU - Stylianou, Yannis

PY - 2000

Y1 - 2000

N2 - We present a new algorithm for adjusting the magnitude spectrum when the fundamental frequency (F0) of a speech signal is altered. The algorithm exploits the correlation between F0 and the magnitude spectrum of speech as represented by line spectral frequencies (LSFs). This correlation is class-dependent, and thus a broad classification of the input is achieved by a Gaussian mixture model (GMM). The within-class dependencies of LSFs on F0 values are captured by constructing their joint probability densities using a series of GMMs, one for each speech class. The proposed system is used for post-processing the pitch modified signal. Perceptual tests showed that the addition of this post-processing system improves the naturalness of the pitch modified signal for large pitch modification factors.

AB - We present a new algorithm for adjusting the magnitude spectrum when the fundamental frequency (F0) of a speech signal is altered. The algorithm exploits the correlation between F0 and the magnitude spectrum of speech as represented by line spectral frequencies (LSFs). This correlation is class-dependent, and thus a broad classification of the input is achieved by a Gaussian mixture model (GMM). The within-class dependencies of LSFs on F0 values are captured by constructing their joint probability densities using a series of GMMs, one for each speech class. The proposed system is used for post-processing the pitch modified signal. Perceptual tests showed that the addition of this post-processing system improves the naturalness of the pitch modified signal for large pitch modification factors.

UR - http://www.scopus.com/inward/record.url?scp=0033693289&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033693289&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0033693289

VL - 2

SP - 949

EP - 952

BT - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

PB - IEEE

ER -