Back to search

Institutional Repository of Vilnius University

Vokalų ištraukimas iš muzikinių failų /

Abstract

dc:description

This thesis examines the problem of vocal extraction from music recordings using artificial intelligence­based models. The aim of the work is to design, train, and compare several neural ne­ twork architectures for separating the vocal signal from the accompaniment. The thesis discusses several neural network architectures: U­net, BS­-RoFormer, MB-­RoFormer, ”Windowed-­sink” MB-­RoFormer, and Mamba blocks. Experiments were conducted with four architecture groups with variations: U­net with trans­ former blocks (three variations), ”Windowed-­sink” MB-­RoFormer, MB­-RoFormer (two variations), and MB-­RoFormer with Mamba blocks (two variations). The models were trained using the MUS­DB18 dataset and evaluated according to the SDR, SI-­SDR, SIR, SAR, and ISR metrics.

Degree

thesis:*
Grantor dc:publisher
Institutional Repository of Vilnius University
Year dc:date
2026

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Belevičius, Benediktas,

Rights

dc:rights
Statement dc:rights
  • info:eu-repo/semantics/openAccess
Language dc:language
lit

Identifiers

dc:identifier.*

Chain of custody

source
Harvested from
Vilnius University
Base URL
epublications.vu.lt/oai
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
related terms
citation

Belevičius, Benediktas,. Vokalų ištraukimas iš muzikinių failų /. Institutional Repository of Vilnius University, 2026. https://repository.vu.lt/VU:ELABAETD308337124&prefLang=en_US