Advanced

Packet loss concealment with natural variations using HMM

Murthi, Manohar N. ; Rodbro, Christoffer A. ; Andersen, Sören Vang LU and Jensen, Soren Holdt (2006) IEEE International Conference on Acoustics, Speech, and Signal Processing, 2006 p.21-24
Abstract
Packet loss concealment (PLC) at a receiver has a substantial effect on the speech quality in Voice over IP. Most conventional PLC systems have largely relied upon variations of signal repetition and overlap-add interpolation which can produce speech signals that do not follow the larger overall statistical trends. In this paper, we demonstrate how Hidden Markov Models can be utilized to effect PLC based on statistical signal processing. In particular, we show how HMM-based PLC yields conditional density functions that can be utilized by various statistical estimation methods that produce signal parameter estimates that produce more natural variation than conventional PLC methods, thereby providing much better speech quality.
Please use this url to cite or link to this publication:
author
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
host publication
2006 IEEE International Conference on Acoustics, Speech and Signal Processing, Vols 1-13
pages
21 - 24
publisher
IEEE - Institute of Electrical and Electronics Engineers Inc.
conference name
IEEE International Conference on Acoustics, Speech, and Signal Processing, 2006
conference location
Toulouse, France
conference dates
2006-05-14 - 2006-05-19
external identifiers
  • wos:000245559900006
  • scopus:33947664563
ISSN
1520-6149
language
English
LU publication?
no
id
a5c39be5-88cb-4f88-a36d-e6e87a264e1f (old id 4092521)
date added to LUP
2016-04-01 17:14:25
date last changed
2020-01-12 20:11:38
@inproceedings{a5c39be5-88cb-4f88-a36d-e6e87a264e1f,
  abstract     = {Packet loss concealment (PLC) at a receiver has a substantial effect on the speech quality in Voice over IP. Most conventional PLC systems have largely relied upon variations of signal repetition and overlap-add interpolation which can produce speech signals that do not follow the larger overall statistical trends. In this paper, we demonstrate how Hidden Markov Models can be utilized to effect PLC based on statistical signal processing. In particular, we show how HMM-based PLC yields conditional density functions that can be utilized by various statistical estimation methods that produce signal parameter estimates that produce more natural variation than conventional PLC methods, thereby providing much better speech quality.},
  author       = {Murthi, Manohar N. and Rodbro, Christoffer A. and Andersen, Sören Vang and Jensen, Soren Holdt},
  booktitle    = {2006 IEEE International Conference on Acoustics, Speech and Signal Processing, Vols 1-13},
  issn         = {1520-6149},
  language     = {eng},
  pages        = {21--24},
  publisher    = {IEEE - Institute of Electrical and Electronics Engineers Inc.},
  title        = {Packet loss concealment with natural variations using HMM},
  year         = {2006},
}