Internship report - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Master Thesis Year : 2013

Internship report


Matching pursuit (MP), particularly using the Gammatones dic- tionary, has become a popular tool in sparse representations of speech/audio signals. The classical MP algorithm does not however take into account psychoacoustical aspects of the audi- tory system. Recently two algorithms, called PAMP and PMP have been introduced in order to select only perceptually rele- vant atoms during MP decomposition. In this paper we compare this two algorithms on few speech sentences. The results sug- gest that PMP, which also has the strong advantage of including an implicit stop criterion, always outperforms PAMP as well as classical MP. We then raise the question of whether the Gam- matones dictionary is the best choice when using PMP. We thus compare it to the popular Gabor and damped-Sinusoids dictio- naries. The results suggest that Gammatones always outperform damped-Sinusoids, and that Gabor yield better reconstruction quality but with higher atoms rate.
Fichier principal
Vignette du fichier
Internship_report-VINUESA.pdf (1.54 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-00830202 , version 1 (25-08-2013)


  • HAL Id : hal-00830202 , version 1


Nicolas Vinuesa. Internship report. Signal and Image Processing. 2013. ⟨hal-00830202⟩


333 View
1540 Download


Gmail Facebook X LinkedIn More