Nonlinear long-term prediction of speech based on truncated Volterra series

Vladimir Despotovic; Norbert Goertz; Zoran Peric

doi:10.1109/TASL.2011.2169788

Nonlinear long-term prediction of speech based on truncated Volterra series

Vladimir Despotovic^*
, Norbert Goertz
, Zoran Peric

^*Corresponding author for this work

Research output: Contribution to journal › Article › Research › peer-review

17 Citations (Scopus)

Abstract

Previous studies of nonlinear prediction of speech have been mostly focused on short-term prediction. This paper presents long-term nonlinear prediction based on second-order Volterra filters. It will be shown that the presented predictor can outperform conventional linear prediction techniques in terms of prediction gain and "whiter" residuals.

Original language	English
Article number	2169788
Pages (from-to)	1069-1073
Number of pages	5
Journal	IEEE Transactions on Audio, Speech and Language Processing
Volume	20
Issue number	3
DOIs	https://doi.org/10.1109/TASL.2011.2169788
Publication status	Published - 2012
Externally published	Yes

Keywords

Linear predictive coding
Long-term prediction
Nonlinear filters
Volterra series

Access to Document

10.1109/TASL.2011.2169788

Cite this

@article{9f66796e038640068272c9d399dbad1b,

title = "Nonlinear long-term prediction of speech based on truncated Volterra series",

abstract = "Previous studies of nonlinear prediction of speech have been mostly focused on short-term prediction. This paper presents long-term nonlinear prediction based on second-order Volterra filters. It will be shown that the presented predictor can outperform conventional linear prediction techniques in terms of prediction gain and {"}whiter{"} residuals.",

keywords = "Linear predictive coding, Long-term prediction, Nonlinear filters, Volterra series",

author = "Vladimir Despotovic and Norbert Goertz and Zoran Peric",

year = "2012",

doi = "10.1109/TASL.2011.2169788",

language = "English",

volume = "20",

pages = "1069--1073",

journal = "IEEE Transactions on Audio, Speech and Language Processing",

issn = "1558-7916",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "3",

}

TY - JOUR

T1 - Nonlinear long-term prediction of speech based on truncated Volterra series

AU - Despotovic, Vladimir

AU - Goertz, Norbert

AU - Peric, Zoran

PY - 2012

Y1 - 2012

N2 - Previous studies of nonlinear prediction of speech have been mostly focused on short-term prediction. This paper presents long-term nonlinear prediction based on second-order Volterra filters. It will be shown that the presented predictor can outperform conventional linear prediction techniques in terms of prediction gain and "whiter" residuals.

AB - Previous studies of nonlinear prediction of speech have been mostly focused on short-term prediction. This paper presents long-term nonlinear prediction based on second-order Volterra filters. It will be shown that the presented predictor can outperform conventional linear prediction techniques in terms of prediction gain and "whiter" residuals.

KW - Linear predictive coding

KW - Long-term prediction

KW - Nonlinear filters

KW - Volterra series

UR - https://www.scopus.com/pages/publications/84857349972

U2 - 10.1109/TASL.2011.2169788

DO - 10.1109/TASL.2011.2169788

M3 - Article

AN - SCOPUS:84857349972

SN - 1558-7916

VL - 20

SP - 1069

EP - 1073

JO - IEEE Transactions on Audio, Speech and Language Processing

JF - IEEE Transactions on Audio, Speech and Language Processing

IS - 3

M1 - 2169788

ER -

Nonlinear long-term prediction of speech based on truncated Volterra series

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this