On the computation of the Kullback-Leibler measure for spectral distances

Raymond N.J. Veldhuis; E. Klabbers

doi:10.1109/TSA.2002.805641

On the computation of the Kullback-Leibler measure for spectral distances

Raymond N.J. Veldhuis, E. Klabbers

Research output: Contribution to journal › Article › Academic › peer-review

23 Citations (Scopus)

184 Downloads (Pure)

Abstract

Efficient algorithms for the exact and approximate computation of the symmetrical Kullback-Leibler (1998) measure for spectral distances are presented for linear predictive coding (LPC) spectra. An interpretation of this measure is given in terms of the poles of the spectra. The performances of the algorithms in terms of accuracy and computational complexity are assessed for the application of computing concatenation costs in unit-selection-based speech synthesis. With the same complexity and storage requirements, the exact method is superior in terms of accuracy.

Original language	Undefined
Pages (from-to)	100-103
Number of pages	4
Journal	IEEE transactions on speech and audio processing
Volume	11
Issue number	1
DOIs	https://doi.org/10.1109/TSA.2002.805641
Publication status	Published - Jan 2003

Keywords

EWI-807
IR-45311
METIS-211893

Access to Document

10.1109/TSA.2002.805641

01179384

http://doi.ieeecomputersociety.org/10.1109/TSA.2002.805641

Cite this

@article{2df6590ae51b437b93df1461f748908a,

title = "On the computation of the Kullback-Leibler measure for spectral distances",

abstract = "Efficient algorithms for the exact and approximate computation of the symmetrical Kullback-Leibler (1998) measure for spectral distances are presented for linear predictive coding (LPC) spectra. An interpretation of this measure is given in terms of the poles of the spectra. The performances of the algorithms in terms of accuracy and computational complexity are assessed for the application of computing concatenation costs in unit-selection-based speech synthesis. With the same complexity and storage requirements, the exact method is superior in terms of accuracy.",

keywords = "EWI-807, IR-45311, METIS-211893",

author = "Veldhuis, {Raymond N.J.} and E. Klabbers",

note = "Imported from DIES",

year = "2003",

month = jan,

doi = "10.1109/TSA.2002.805641",

language = "Undefined",

volume = "11",

pages = "100--103",

journal = "IEEE transactions on speech and audio processing",

issn = "1063-6676",

publisher = "IEEE",

number = "1",

}

TY - JOUR

T1 - On the computation of the Kullback-Leibler measure for spectral distances

AU - Veldhuis, Raymond N.J.

AU - Klabbers, E.

N1 - Imported from DIES

PY - 2003/1

Y1 - 2003/1

N2 - Efficient algorithms for the exact and approximate computation of the symmetrical Kullback-Leibler (1998) measure for spectral distances are presented for linear predictive coding (LPC) spectra. An interpretation of this measure is given in terms of the poles of the spectra. The performances of the algorithms in terms of accuracy and computational complexity are assessed for the application of computing concatenation costs in unit-selection-based speech synthesis. With the same complexity and storage requirements, the exact method is superior in terms of accuracy.

AB - Efficient algorithms for the exact and approximate computation of the symmetrical Kullback-Leibler (1998) measure for spectral distances are presented for linear predictive coding (LPC) spectra. An interpretation of this measure is given in terms of the poles of the spectra. The performances of the algorithms in terms of accuracy and computational complexity are assessed for the application of computing concatenation costs in unit-selection-based speech synthesis. With the same complexity and storage requirements, the exact method is superior in terms of accuracy.

KW - EWI-807

KW - IR-45311

KW - METIS-211893

U2 - 10.1109/TSA.2002.805641

DO - 10.1109/TSA.2002.805641

M3 - Article

SN - 1063-6676

VL - 11

SP - 100

EP - 103

JO - IEEE transactions on speech and audio processing

JF - IEEE transactions on speech and audio processing

IS - 1

ER -