Hidden Markov model-based packet loss concealment for voice over IP

Christoffer A. Rødbro, Manohar Murthi, Søren Vang Andersen, Søren Holdt Jensen

Research output: Contribution to journalArticle

49 Citations (Scopus)

Abstract

As voice over IP proliferates, packet loss concealment (PLC) at the receiver has emerged as an important factor in determining voice quality of service. Through the use of heuristic variations of signal and parameter repetition and overlap-add interpolation to handle packet loss, conventional PLC systems largely ignore the dynamics of the statistical evolution of the speech signal, possibly leading to perceptually annoying artifacts. To address this problem, we propose the use of hidden Markov models for PLC. With a hidden Markov model (HMM) tracking the evolution of speech signal parameters, we demonstrate how PLC is performed within a statistical signal processing framework. Moreover, we show how the HMM is used to index a specially designed PLC module for the particular signal context, leading to signal-contingent PLC. Simulation examples, objective tests, and subjective listening tests are provided showing the ability of an HMM-based PLC built with a sinusoidal analysis/synthesis model to provide better loss concealment than a conventional PLC based on the same sinusoidal model for all types of speech signals, including onsets and signal transitions.

Original languageEnglish
Pages (from-to)1609-1623
Number of pages15
JournalIEEE Transactions on Audio, Speech and Language Processing
Volume14
Issue number5
DOIs
StatePublished - Sep 1 2006

Fingerprint

Hidden Markov models
Packet loss
Interpolation
Quality of service
Signal processing
interpolation
artifacts
signal processing
repetition
receivers
modules
synthesis

Keywords

  • Hidden Markov models (HMMs)
  • Internet telephony
  • Packet loss
  • Packet loss concealment
  • Packet switching
  • Speech coding
  • Speech communication
  • Voice over IP

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Acoustics and Ultrasonics

Cite this

Hidden Markov model-based packet loss concealment for voice over IP. / Rødbro, Christoffer A.; Murthi, Manohar; Andersen, Søren Vang; Jensen, Søren Holdt.

In: IEEE Transactions on Audio, Speech and Language Processing, Vol. 14, No. 5, 01.09.2006, p. 1609-1623.

Research output: Contribution to journalArticle

Rødbro, Christoffer A. ; Murthi, Manohar ; Andersen, Søren Vang ; Jensen, Søren Holdt. / Hidden Markov model-based packet loss concealment for voice over IP. In: IEEE Transactions on Audio, Speech and Language Processing. 2006 ; Vol. 14, No. 5. pp. 1609-1623.
@article{a2f5db790ed745e4825410e3b3e91240,
title = "Hidden Markov model-based packet loss concealment for voice over IP",
abstract = "As voice over IP proliferates, packet loss concealment (PLC) at the receiver has emerged as an important factor in determining voice quality of service. Through the use of heuristic variations of signal and parameter repetition and overlap-add interpolation to handle packet loss, conventional PLC systems largely ignore the dynamics of the statistical evolution of the speech signal, possibly leading to perceptually annoying artifacts. To address this problem, we propose the use of hidden Markov models for PLC. With a hidden Markov model (HMM) tracking the evolution of speech signal parameters, we demonstrate how PLC is performed within a statistical signal processing framework. Moreover, we show how the HMM is used to index a specially designed PLC module for the particular signal context, leading to signal-contingent PLC. Simulation examples, objective tests, and subjective listening tests are provided showing the ability of an HMM-based PLC built with a sinusoidal analysis/synthesis model to provide better loss concealment than a conventional PLC based on the same sinusoidal model for all types of speech signals, including onsets and signal transitions.",
keywords = "Hidden Markov models (HMMs), Internet telephony, Packet loss, Packet loss concealment, Packet switching, Speech coding, Speech communication, Voice over IP",
author = "R{\o}dbro, {Christoffer A.} and Manohar Murthi and Andersen, {S{\o}ren Vang} and Jensen, {S{\o}ren Holdt}",
year = "2006",
month = "9",
day = "1",
doi = "10.1109/TSA.2005.858561",
language = "English",
volume = "14",
pages = "1609--1623",
journal = "IEEE Transactions on Speech and Audio Processing",
issn = "1558-7916",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "5",

}

TY - JOUR

T1 - Hidden Markov model-based packet loss concealment for voice over IP

AU - Rødbro, Christoffer A.

AU - Murthi, Manohar

AU - Andersen, Søren Vang

AU - Jensen, Søren Holdt

PY - 2006/9/1

Y1 - 2006/9/1

N2 - As voice over IP proliferates, packet loss concealment (PLC) at the receiver has emerged as an important factor in determining voice quality of service. Through the use of heuristic variations of signal and parameter repetition and overlap-add interpolation to handle packet loss, conventional PLC systems largely ignore the dynamics of the statistical evolution of the speech signal, possibly leading to perceptually annoying artifacts. To address this problem, we propose the use of hidden Markov models for PLC. With a hidden Markov model (HMM) tracking the evolution of speech signal parameters, we demonstrate how PLC is performed within a statistical signal processing framework. Moreover, we show how the HMM is used to index a specially designed PLC module for the particular signal context, leading to signal-contingent PLC. Simulation examples, objective tests, and subjective listening tests are provided showing the ability of an HMM-based PLC built with a sinusoidal analysis/synthesis model to provide better loss concealment than a conventional PLC based on the same sinusoidal model for all types of speech signals, including onsets and signal transitions.

AB - As voice over IP proliferates, packet loss concealment (PLC) at the receiver has emerged as an important factor in determining voice quality of service. Through the use of heuristic variations of signal and parameter repetition and overlap-add interpolation to handle packet loss, conventional PLC systems largely ignore the dynamics of the statistical evolution of the speech signal, possibly leading to perceptually annoying artifacts. To address this problem, we propose the use of hidden Markov models for PLC. With a hidden Markov model (HMM) tracking the evolution of speech signal parameters, we demonstrate how PLC is performed within a statistical signal processing framework. Moreover, we show how the HMM is used to index a specially designed PLC module for the particular signal context, leading to signal-contingent PLC. Simulation examples, objective tests, and subjective listening tests are provided showing the ability of an HMM-based PLC built with a sinusoidal analysis/synthesis model to provide better loss concealment than a conventional PLC based on the same sinusoidal model for all types of speech signals, including onsets and signal transitions.

KW - Hidden Markov models (HMMs)

KW - Internet telephony

KW - Packet loss

KW - Packet loss concealment

KW - Packet switching

KW - Speech coding

KW - Speech communication

KW - Voice over IP

UR - http://www.scopus.com/inward/record.url?scp=33744969796&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33744969796&partnerID=8YFLogxK

U2 - 10.1109/TSA.2005.858561

DO - 10.1109/TSA.2005.858561

M3 - Article

AN - SCOPUS:33744969796

VL - 14

SP - 1609

EP - 1623

JO - IEEE Transactions on Speech and Audio Processing

JF - IEEE Transactions on Speech and Audio Processing

SN - 1558-7916

IS - 5

ER -