Mercurial > hg > smallbox

comparison Problems/AudioDeclipping_reconstruct.m @ 161:f42aa8bcb82f ivand_dev

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
debug and clean the SMALLbox Problems code
author Ivan Damnjanovic lnx <ivan.damnjanovic@eecs.qmul.ac.uk>
date Wed, 31 Aug 2011 12:02:19 +0100
parents 31d2864dfdd4
children
comparison
equal deleted inserted replaced
155:b14209313ba4 161:f42aa8bcb82f
1 function reconstructed=AudioDeclipping_reconstruct(y, Problem, SparseDict) 1 function reconstructed = AudioDeclipping_reconstruct(y, Problem)
2 %% Audio declipping Problem reconstruction function 2 %% Audio declipping Problem reconstruction function
3 % 3 %
4 % This reconstruction function is using sparse representation y 4 % This reconstruction function is using sparse representation y
5 % in dictionary Problem.A to reconstruct declipped audio. 5 % in dictionary Problem.A to reconstruct declipped audio.
6 % The output structure has following fields:
7 % audioAllSamples - signal with all samples taken from reconstructed
8 % signal
9 % audioOnlyClipped - only clipped samples are reconstructed,
10 % others are taken from original signal
11 % snrAll - psnr of whole signal
12 % snrMiss - psnr of the reconstructed clipped samples
6 % 13 %
7 % [1] I. Damnjanovic, M. E. P. Davies, and M. P. Plumbley "SMALLbox - an 14 % [1] I. Damnjanovic, M. E. P. Davies, and M. P. Plumbley "SMALLbox - an
8 % evaluation framework for sparse representations and dictionary 15 % evaluation framework for sparse representations and dictionary
9 % learning algorithms," V. Vigneron et al. (Eds.): LVA/ICA 2010, 16 % learning algorithms," V. Vigneron et al. (Eds.): LVA/ICA 2010,
10 % Springer-Verlag, Berlin, Germany, LNCS 6365, pp. 418-425 17 % Springer-Verlag, Berlin, Germany, LNCS 6365, pp. 418-425