Retrieving Sparse Patterns Using a Compressed Sensing Framework: Applications to Speech Coding Based on Sparse Linear Prediction

Daniele Giacobello, Søren Holdt Jensen, Mads Græsbøll Christensen, Manohar Murthi, Marc Moonen

Research output: Contribution to journalArticle

67 Citations (Scopus)

Abstract

Encouraged by the promising application of compressed sensing in signal compression, we investigate its formulation and application in the context of speech coding based on sparse linear prediction. In particular, a compressed sensing method can be devised to compute a sparse approximation of speech in the residual domain when sparse linear prediction is involved. We compare the method of computing a sparse prediction residual with the optimal technique based on an exhaustive search of the possible nonzero locations and the well known Multi-Pulse Excitation, the first encoding technique to introduce the sparsity concept in speech coding. Experimental results demonstrate the potential of compressed sensing in speech coding techniques, offering high perceptual quality with a very sparse approximated prediction residual.

Original languageEnglish (US)
Pages (from-to)103-106
Number of pages4
JournalIEEE Signal Processing Letters
Volume17
Issue number1
DOIs
StatePublished - 2010

Fingerprint

Speech Coding
Speech coding
Compressed sensing
Linear Prediction
Compressed Sensing
Sparse Approximation
Prediction
Exhaustive Search
Sparsity
Encoding
Compression
Excitation
Framework
Formulation
Computing
Experimental Results
Demonstrate

Keywords

  • compressed sensing
  • Compressive sampling
  • sparse approximation
  • speech analysis
  • speech coding

ASJC Scopus subject areas

  • Signal Processing
  • Applied Mathematics
  • Electrical and Electronic Engineering

Cite this

Retrieving Sparse Patterns Using a Compressed Sensing Framework : Applications to Speech Coding Based on Sparse Linear Prediction. / Giacobello, Daniele; Jensen, Søren Holdt; Christensen, Mads Græsbøll; Murthi, Manohar; Moonen, Marc.

In: IEEE Signal Processing Letters, Vol. 17, No. 1, 2010, p. 103-106.

Research output: Contribution to journalArticle

Giacobello, Daniele ; Jensen, Søren Holdt ; Christensen, Mads Græsbøll ; Murthi, Manohar ; Moonen, Marc. / Retrieving Sparse Patterns Using a Compressed Sensing Framework : Applications to Speech Coding Based on Sparse Linear Prediction. In: IEEE Signal Processing Letters. 2010 ; Vol. 17, No. 1. pp. 103-106.
@article{bc2d157b2e274096ac366abcdc35bceb,
title = "Retrieving Sparse Patterns Using a Compressed Sensing Framework: Applications to Speech Coding Based on Sparse Linear Prediction",
abstract = "Encouraged by the promising application of compressed sensing in signal compression, we investigate its formulation and application in the context of speech coding based on sparse linear prediction. In particular, a compressed sensing method can be devised to compute a sparse approximation of speech in the residual domain when sparse linear prediction is involved. We compare the method of computing a sparse prediction residual with the optimal technique based on an exhaustive search of the possible nonzero locations and the well known Multi-Pulse Excitation, the first encoding technique to introduce the sparsity concept in speech coding. Experimental results demonstrate the potential of compressed sensing in speech coding techniques, offering high perceptual quality with a very sparse approximated prediction residual.",
keywords = "compressed sensing, Compressive sampling, sparse approximation, speech analysis, speech coding",
author = "Daniele Giacobello and Jensen, {S{\o}ren Holdt} and Christensen, {Mads Gr{\ae}sb{\o}ll} and Manohar Murthi and Marc Moonen",
year = "2010",
doi = "10.1109/LSP.2009.2034560",
language = "English (US)",
volume = "17",
pages = "103--106",
journal = "IEEE Signal Processing Letters",
issn = "1070-9908",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "1",

}

TY - JOUR

T1 - Retrieving Sparse Patterns Using a Compressed Sensing Framework

T2 - Applications to Speech Coding Based on Sparse Linear Prediction

AU - Giacobello, Daniele

AU - Jensen, Søren Holdt

AU - Christensen, Mads Græsbøll

AU - Murthi, Manohar

AU - Moonen, Marc

PY - 2010

Y1 - 2010

N2 - Encouraged by the promising application of compressed sensing in signal compression, we investigate its formulation and application in the context of speech coding based on sparse linear prediction. In particular, a compressed sensing method can be devised to compute a sparse approximation of speech in the residual domain when sparse linear prediction is involved. We compare the method of computing a sparse prediction residual with the optimal technique based on an exhaustive search of the possible nonzero locations and the well known Multi-Pulse Excitation, the first encoding technique to introduce the sparsity concept in speech coding. Experimental results demonstrate the potential of compressed sensing in speech coding techniques, offering high perceptual quality with a very sparse approximated prediction residual.

AB - Encouraged by the promising application of compressed sensing in signal compression, we investigate its formulation and application in the context of speech coding based on sparse linear prediction. In particular, a compressed sensing method can be devised to compute a sparse approximation of speech in the residual domain when sparse linear prediction is involved. We compare the method of computing a sparse prediction residual with the optimal technique based on an exhaustive search of the possible nonzero locations and the well known Multi-Pulse Excitation, the first encoding technique to introduce the sparsity concept in speech coding. Experimental results demonstrate the potential of compressed sensing in speech coding techniques, offering high perceptual quality with a very sparse approximated prediction residual.

KW - compressed sensing

KW - Compressive sampling

KW - sparse approximation

KW - speech analysis

KW - speech coding

UR - http://www.scopus.com/inward/record.url?scp=85008023999&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85008023999&partnerID=8YFLogxK

U2 - 10.1109/LSP.2009.2034560

DO - 10.1109/LSP.2009.2034560

M3 - Article

AN - SCOPUS:85008023999

VL - 17

SP - 103

EP - 106

JO - IEEE Signal Processing Letters

JF - IEEE Signal Processing Letters

SN - 1070-9908

IS - 1

ER -