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 belter speech quality.
| Original language | English |
|---|---|
| Title of host publication | ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings |
| Volume | 1 |
| State | Published - Dec 1 2006 |
| Event | 2006 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2006 - Toulouse, France Duration: May 14 2006 → May 19 2006 |
Other
| Other | 2006 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2006 |
|---|---|
| Country | France |
| City | Toulouse |
| Period | 5/14/06 → 5/19/06 |
Fingerprint
ASJC Scopus subject areas
- Electrical and Electronic Engineering
- Signal Processing
- Acoustics and Ultrasonics
Cite this
Packet loss concealment with natural variations using HMM. / Murthi, Manohar; Rødbro, Christoffer A.; Andersen, Søren Vang; Jensen, Søren Holdt.
ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Vol. 1 2006. 1659947.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Packet loss concealment with natural variations using HMM
AU - Murthi, Manohar
AU - Rødbro, Christoffer A.
AU - Andersen, Søren Vang
AU - Jensen, Søren Holdt
PY - 2006/12/1
Y1 - 2006/12/1
N2 - 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 belter speech quality.
AB - 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 belter speech quality.
UR - http://www.scopus.com/inward/record.url?scp=33947664563&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=33947664563&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:33947664563
SN - 142440469X
SN - 9781424404698
VL - 1
BT - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
ER -