Psychoacoustically motivated audio declipper based on sparsity
This software is devoted to the restoration of audio signals degraded by clipping. The sparsity of the processed signal in the DGT is utilized. To achieve an improved perceptual quality of the restoration, weighting of the coefficients in l1-minimization based on global masking threshold, absolute threshold of hearing and a simple parabola is used. The Douglas-Rachford algorithm is used as the optimization algorithm. The implementation is done in MATLAB.
declipping; sparse recovery; psychoacoustics; masking; sparsity
pracovna SD5.66, Technická 12, 61600 Brno