Robust low-delay audio coding using multiple descriptions

Gerald Schuller, Jelena Kovacevic, Francois Masson, Vivek K. Goyal

Research output: Contribution to journalArticle

Abstract

This paper proposes an encoding method for high-quality, low-delay audio communication that is robust to losses in packetized transmission. Robustness is provided by a multiple description vector quantization (MDVQ) technique that is designed to minimize the mean-squared error (MSE). The key to applying this technique effectively is the use of psycho-acoustically controlled pre- and post-filters that make the mean-squared quantization error perceptually relevant. Experiments show that the MDVQ-based encoder yields better results - in both MSE and subjective audio quality - than simple alternative coders with the same low delay.

Original languageEnglish (US)
Pages (from-to)1014-1023
Number of pages10
JournalIEEE Transactions on Speech and Audio Processing
Volume13
Issue number5
DOIs
StatePublished - Sep 1 2005

Fingerprint

coding
vector quantization
Vector quantization
coders
filters
communication
Communication
Experiments

Keywords

  • Audio coding
  • Error robustness
  • LMS
  • Low delay
  • Multiple descriptions
  • Predictive coding
  • Psycho-acoustics

ASJC Scopus subject areas

  • Software
  • Acoustics and Ultrasonics
  • Computer Vision and Pattern Recognition
  • Electrical and Electronic Engineering

Cite this

Robust low-delay audio coding using multiple descriptions. / Schuller, Gerald; Kovacevic, Jelena; Masson, Francois; Goyal, Vivek K.

In: IEEE Transactions on Speech and Audio Processing, Vol. 13, No. 5, 01.09.2005, p. 1014-1023.

Research output: Contribution to journalArticle

Schuller, Gerald ; Kovacevic, Jelena ; Masson, Francois ; Goyal, Vivek K. / Robust low-delay audio coding using multiple descriptions. In: IEEE Transactions on Speech and Audio Processing. 2005 ; Vol. 13, No. 5. pp. 1014-1023.
@article{03762a2303b343b2adc3b5906570c9f0,
title = "Robust low-delay audio coding using multiple descriptions",
abstract = "This paper proposes an encoding method for high-quality, low-delay audio communication that is robust to losses in packetized transmission. Robustness is provided by a multiple description vector quantization (MDVQ) technique that is designed to minimize the mean-squared error (MSE). The key to applying this technique effectively is the use of psycho-acoustically controlled pre- and post-filters that make the mean-squared quantization error perceptually relevant. Experiments show that the MDVQ-based encoder yields better results - in both MSE and subjective audio quality - than simple alternative coders with the same low delay.",
keywords = "Audio coding, Error robustness, LMS, Low delay, Multiple descriptions, Predictive coding, Psycho-acoustics",
author = "Gerald Schuller and Jelena Kovacevic and Francois Masson and Goyal, {Vivek K.}",
year = "2005",
month = "9",
day = "1",
doi = "10.1109/TSA.2005.853205",
language = "English (US)",
volume = "13",
pages = "1014--1023",
journal = "IEEE Transactions on Speech and Audio Processing",
issn = "1063-6676",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "5",

}

TY - JOUR

T1 - Robust low-delay audio coding using multiple descriptions

AU - Schuller, Gerald

AU - Kovacevic, Jelena

AU - Masson, Francois

AU - Goyal, Vivek K.

PY - 2005/9/1

Y1 - 2005/9/1

N2 - This paper proposes an encoding method for high-quality, low-delay audio communication that is robust to losses in packetized transmission. Robustness is provided by a multiple description vector quantization (MDVQ) technique that is designed to minimize the mean-squared error (MSE). The key to applying this technique effectively is the use of psycho-acoustically controlled pre- and post-filters that make the mean-squared quantization error perceptually relevant. Experiments show that the MDVQ-based encoder yields better results - in both MSE and subjective audio quality - than simple alternative coders with the same low delay.

AB - This paper proposes an encoding method for high-quality, low-delay audio communication that is robust to losses in packetized transmission. Robustness is provided by a multiple description vector quantization (MDVQ) technique that is designed to minimize the mean-squared error (MSE). The key to applying this technique effectively is the use of psycho-acoustically controlled pre- and post-filters that make the mean-squared quantization error perceptually relevant. Experiments show that the MDVQ-based encoder yields better results - in both MSE and subjective audio quality - than simple alternative coders with the same low delay.

KW - Audio coding

KW - Error robustness

KW - LMS

KW - Low delay

KW - Multiple descriptions

KW - Predictive coding

KW - Psycho-acoustics

UR - http://www.scopus.com/inward/record.url?scp=27644526862&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=27644526862&partnerID=8YFLogxK

U2 - 10.1109/TSA.2005.853205

DO - 10.1109/TSA.2005.853205

M3 - Article

AN - SCOPUS:27644526862

VL - 13

SP - 1014

EP - 1023

JO - IEEE Transactions on Speech and Audio Processing

JF - IEEE Transactions on Speech and Audio Processing

SN - 1063-6676

IS - 5

ER -