Automatsko generiranje glazbe temeljeno na umjetnim neuronskim mrežama

master's thesis

• Title (english): Automatic Generation of Music Based on Artifical Neural Networks
• Author: Domagoj Pereglin
• Mentor: Marko Čupić (mentor)
• Committee member: Marko Čupić (predsjednik povjerenstva)
• Committee member: Marin Golub (član povjerenstva)
• Committee member: Jan Šnajder (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2017-07-07, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Korištenje algoritama umjetne inteligencije u skladanju glazbe, tj. algoritamsko skladanje, djelomična je ili potpuna automatizacija procesa skladanja glazbe upotrebom računala. Od pedesetih godina prošlog stoljeća koristile su se različite tehnike umjetne inteligencije u cilju ostvarenja automatizacije tog procesa. U ovom radu objašnjene su umjetne neuronske mreže. Poseban je naglasak stavljen na povratne neuronske mreže i njihovu primjenu u procesu automatskog generiranja glazbe. Opisana je implementacija sustava automatskog generiranja glazbe temeljena na mrežama LSTM. Opisani su rezultati generiranja te su dane poveznice na kod i slušne primjere.

Abstract (english)

Use of artificial intelligence algorithms in music composition, i. e. algorithmic composition, is partially or complete automation of composition process with the help of computers. Since 1950s, many different techniques were used with a goal of automation of that process. In this work artificial neural networks are explained. Special attention is dedicated to recurrent neural networks and their application in a process of automatic generation of music. Implementation of the system based on LSTM networks is described. Generated results are described and links to the code and listening examples are given.

• Keywords: algoritamsko skladanje; povratne neuronske mreže; LSTM; glazba; umjetna inteligencija
• Keywords (english): algorithmic composition; recurrent neural networks; LSTM; music; artificial intelligence
• Language: croatian
• URN:NBN: urn:nbn:hr:168:665153
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-03-13 23:48:55