Kompresija zvuka i zvukovnih datoteka

master's thesis

• Title (english): Compression of sound and sound files
• Author: Karlo Martinek
• Mentor: Goran Igaly (mentor)
• Committee member: Goran Igaly (predsjednik povjerenstva)
• Committee member: Nela Bosner (član povjerenstva)
• Committee member: Vinko Petričević (član povjerenstva)
• Committee member: Marko Tadić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Science; (Department of Mathematics); Zagreb
• Defense date and country: 2020-09-29, Croatia
• Scientific / art field, discipline and subdiscipline: NATURAL SCIENCES; Mathematics

Abstract

Kompresija podataka je proces smanjivanja podataka kako bi se ostvarila ušteda prostora. U ovom radu opisani su postupci digitizacije zvuka, odnosno pretvaranje zvuka u njegovu digitalnu reprezentaciju. Glavni postupci digitizacije su uzorkovanje i kvantizacija. Način kvantiziranja može se modificirati kako bi se postigla ušteda podatkovnog prostora. Postoje dvije glavne vrste kompresije, kompresija s gubitcima i kompresija bez gubitaka. Kompresija s gubitcima u radu je prikazana na primjeru dva MPEG standarda, MPEG-1 i MPEG-2. Oni iskorištavaju psihoakustična svojstva zvuka, poput praga čujnosti ljudskog uha i maskirajućih efekata glasnih tonova, kako bi se postigla visoka kvaliteta zvuka s velikom uštedom prostora. Kompresija bez gubitaka uspijeva smanjiti veličinu zvučnih datoteka, a da se pri tom ne gube informacije. Koriste se posebne metode kodiranja, poput Huffmanovog i Riceovog kodiranja. Kompresija bez gubitaka provodi se u tri glavne faze. U prvoj fazi, signal se dijeli na okvire, zatim se uklanja unutarkanalna korelacija i na kraju se izvršava entropijsko kodiranje. U radu su opisana dva primjera kompresije bez gubitaka FLAC i MPEG-4 ALS.

Abstract (english)

Data compression is the process of encoding information using fewer bits than the original representation. This paper describes the procedures for digitizing sound, ie converting sound into its digital representation. The main digitization procedures are sampling and quantization. The quantization method can be modified to save data space. There are two main types of compression, lossy and lossless compression. Two examples od lossy compression are described in this paper, MPEG-1 and MPEG-2. They use the psychoacoustic properties, such as the hearing threshold and the masking effects of loud tones, to achieve high sound quality with eficient space saving. Lossless compression manages to reduce the size of audio files without losing information. Special coding methods are used, such as Huffman and Rice coding. Lossless compression is performed in three main phases. In the first phase, the signal is divided into frames, then the interchannel correlation is removed and finally the entropy coding is performed. The paper describes two examples of lossless compression - FLAC and MPEG-4 ALS.

• Keywords: kompresija podataka; digitizacija zvuka; MPEG standard; Huffmanovo kodiranje; i Riceovo kodiranje; FLAC; MPEG-4 ALS
• Keywords (english): data compression; digitizing sound; MPEG; Huffman coding; Rice coding; FLAC; MPEG-4 ALS
• Language: croatian
• URN:NBN: urn:nbn:hr:217:291470
• Study programme: Title: Computer Science and Mathematics; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra računarstva i matematike (magistar/magistra računarstva i matematike)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2021-03-01 12:37:25