Perceptual shape VQ of spectral envelope for efficient representation of LPC residual

M. L. Wang, J. F. Yang

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

The authors present an effective spectral envelope (SE) quantisation scheme for parametric speech coders, based on human hearing properties. The variable-dimension SE uniformly sampled vector in frequency is first converted into a fixed, but small, number of nonlinearly spaced frequency bands on the Bark scale. The minimum Bark spectral distortion (BSD) criterion is applied to enable the hearing-based SE vector quantisation (HSEVQ) scheme to quantise the SE vector, achieving a slightly better perceptual quality than the traditional method. A simplified HSEVQ (SSEVQ) scheme is developed by removing some of insensitive functions from the HSEVQ to reduce the complexity of the computation. Simulations reveal that the SSEVQ method reduces the amount of computation of the traditional SE vector quantisation scheme by a factor of nine, while retaining the quality of the reconstructed speech signal.

Original languageEnglish
Pages (from-to)434-442
Number of pages9
JournalIEE Proceedings: Vision, Image and Signal Processing
Volume151
Issue number5
DOIs
Publication statusPublished - 2004 Oct 1

Fingerprint

Vector quantization
Audition
Frequency bands

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

@article{d12e21347db64a6184297fcd9f1ee10b,
title = "Perceptual shape VQ of spectral envelope for efficient representation of LPC residual",
abstract = "The authors present an effective spectral envelope (SE) quantisation scheme for parametric speech coders, based on human hearing properties. The variable-dimension SE uniformly sampled vector in frequency is first converted into a fixed, but small, number of nonlinearly spaced frequency bands on the Bark scale. The minimum Bark spectral distortion (BSD) criterion is applied to enable the hearing-based SE vector quantisation (HSEVQ) scheme to quantise the SE vector, achieving a slightly better perceptual quality than the traditional method. A simplified HSEVQ (SSEVQ) scheme is developed by removing some of insensitive functions from the HSEVQ to reduce the complexity of the computation. Simulations reveal that the SSEVQ method reduces the amount of computation of the traditional SE vector quantisation scheme by a factor of nine, while retaining the quality of the reconstructed speech signal.",
author = "Wang, {M. L.} and Yang, {J. F.}",
year = "2004",
month = "10",
day = "1",
doi = "10.1049/ip-vis:20040809",
language = "English",
volume = "151",
pages = "434--442",
journal = "IEE Proceedings: Vision, Image and Signal Processing",
issn = "1350-245X",
publisher = "Institute of Electrical Engineers",
number = "5",

}

Perceptual shape VQ of spectral envelope for efficient representation of LPC residual. / Wang, M. L.; Yang, J. F.

In: IEE Proceedings: Vision, Image and Signal Processing, Vol. 151, No. 5, 01.10.2004, p. 434-442.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Perceptual shape VQ of spectral envelope for efficient representation of LPC residual

AU - Wang, M. L.

AU - Yang, J. F.

PY - 2004/10/1

Y1 - 2004/10/1

N2 - The authors present an effective spectral envelope (SE) quantisation scheme for parametric speech coders, based on human hearing properties. The variable-dimension SE uniformly sampled vector in frequency is first converted into a fixed, but small, number of nonlinearly spaced frequency bands on the Bark scale. The minimum Bark spectral distortion (BSD) criterion is applied to enable the hearing-based SE vector quantisation (HSEVQ) scheme to quantise the SE vector, achieving a slightly better perceptual quality than the traditional method. A simplified HSEVQ (SSEVQ) scheme is developed by removing some of insensitive functions from the HSEVQ to reduce the complexity of the computation. Simulations reveal that the SSEVQ method reduces the amount of computation of the traditional SE vector quantisation scheme by a factor of nine, while retaining the quality of the reconstructed speech signal.

AB - The authors present an effective spectral envelope (SE) quantisation scheme for parametric speech coders, based on human hearing properties. The variable-dimension SE uniformly sampled vector in frequency is first converted into a fixed, but small, number of nonlinearly spaced frequency bands on the Bark scale. The minimum Bark spectral distortion (BSD) criterion is applied to enable the hearing-based SE vector quantisation (HSEVQ) scheme to quantise the SE vector, achieving a slightly better perceptual quality than the traditional method. A simplified HSEVQ (SSEVQ) scheme is developed by removing some of insensitive functions from the HSEVQ to reduce the complexity of the computation. Simulations reveal that the SSEVQ method reduces the amount of computation of the traditional SE vector quantisation scheme by a factor of nine, while retaining the quality of the reconstructed speech signal.

UR - http://www.scopus.com/inward/record.url?scp=11844286940&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=11844286940&partnerID=8YFLogxK

U2 - 10.1049/ip-vis:20040809

DO - 10.1049/ip-vis:20040809

M3 - Article

AN - SCOPUS:11844286940

VL - 151

SP - 434

EP - 442

JO - IEE Proceedings: Vision, Image and Signal Processing

JF - IEE Proceedings: Vision, Image and Signal Processing

SN - 1350-245X

IS - 5

ER -