Spremenjena diskretna kosinusna transformacija s psihoakustičnim modelom za stiskanje zvoka

diplomsko delo

Grega Premoša (Author), Niko Lukač (Mentor)

Abstract

V diplomskem delu predstavimo metodo za izgubno stiskanje zvoka z uporabo psihoakustičnega modela in algoritma spremenjene diskretne kosinusne transformacije. Podatke prenesemo v frekvenčno domeno in odstranimo odseke, katerih človek ni sposoben zaznati. S kosinusno transformacijo pretvorimo podatke naraščujoče po frekvenci in porežemo visoko frekvenčno vsebino. V zadnji fazi uporabimo stiskanje z Lempel-Ziv algoritmom na podlagi markovske verige (LZMA). V rezultatih smo izvedli analizo občutljivosti glede na različne vrednosti vhodnih parametrov metode ter analizo uspešnosti stiskanja glede na kvaliteto izhoda.

Keywords

stiskanje zvoka;psihoakustični modeli;spremenjena diskretna kosinusna transformacija;frekvenčna analiza;diplomske naloge;

Data

Language:	Slovenian
Year of publishing:	2018
Typology:	2.11 - Undergraduate Thesis
Organization:	UM FERI - Faculty of Electrical Engineering and Computer Science
Publisher:	G. Premoša
UDC:	004.9:512.817(043.2)
COBISS:	21745686
Views:	940
Downloads:	116
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary title:	Sound compression by using modified discrete cosine transform and psychoacoustic model
Secondary abstract:	In our thesis, we present a sound compression algorithm using modified discrete cosine transformation and psychoacoustic model. Original data is transformed into frequency domain, where inaudible parts of the human ear are removed. With modified discrete cosine transform the data is transformed into an ascending list of their frequency content, after which high frequencies are cut for better compression. Final part of the algorithm uses lossless Lempel–Ziv–Markov chain algorithm (LZMA). Implemented algorithm was parameterized and sensitivity analysis was performed on sound quality.
Secondary keywords:	sound compression;psychoacoustic models;modified discrete cosine transform;LZMA;frequency analysis;
URN:	URN:SI:UM:
Type (COBISS):	Bachelor thesis/paper
Thesis comment:	Univ. v Mariboru, Fak. za elektrotehniko, računalništvo in informatiko, Računalništvo in informacijske tehnologije
Pages:	VIII, 39 f.
ID:	10944658