izpis_h1_title_alt

Iskanje po zbirkah ljudske glasbe na podlagi mrmranja
ID MITTONI, TADEJ (Author), ID Kavčič, Alenka (Mentor) More about this mentor... This link opens in a new window, ID Marolt, Matija (Co-mentor)

.pdfPDF - Presentation file, Download (1,55 MB)
MD5: 31EE7E75CE3CFA0388E3AA56BAC67573
PID: 20.500.12556/rul/8af43488-b2f2-413f-b357-9c5488d9cd2f

Abstract
Glavna naloga sistemov QBH je identifikacija vhodnemu zamrmranemu vzorcu najbolj podobnih glasbenih del v podatkovni bazi. Postopek se začne z zajemom zamrmrane melodije na strani uporabnika, nadaljuje s transkripcijo zamrmranega vzorca, nato sistem s pomočjo algoritmov za iskanje podobnosti najde določeno število najbolj podobnih glasbenih del v podatkovni bazi, kar je sporočeno nazaj uporabniku. Cilj diplomskega dela je preučitev že ponujenih sistemov QBH ter pozneje implementacija najoptimalnejšega za uporabo v spletni aplikaciji EtnoFletno. Za fazo transkripcije zamrmranega vzorca sta bila preizkušena algoritma YIN ter njegova izboljšana različica pYIN. Preizkušeni algoritmi za iskanje podobnosti so bili DTW, razdalja urejanja, Spring ter SMGT in SMBGT. Zaradi cilja po optimizaciji tako transkripcijskih kot tudi iskalnih algoritmov so bili vsi poskusi izvedeni s podatkovno bazo, ki je vsebovala slovenske ljudske pesmi. Končni rezultati so pokazali, da je najoptimalnejši sistem QBH sestavljen iz transkripcijskega algoritma pYIN ter iskalnega algoritma SMBGT, ki lahko glede na pevske sposobnosti uporabnika preklaplja med dvema sklopoma parametrov.

Language:Slovenian
Keywords:poizvedba z mrmranjem, transkripcija zamrmranega vzorca, EtnoFletno, slovenske ljudske pesmi, algoritmi za iskanje najbolj podobnih glasbenih del, algoritmi za detekcijo višine tona, dinamično programiranje, verjetnostni YIN, ujemanje podvzorca
Work type:Bachelor thesis/paper
Organization:FRI - Faculty of Computer and Information Science
Year:2016
PID:20.500.12556/RUL-80814 This link opens in a new window
Publication date in RUL:03.03.2016
Views:1519
Downloads:585
Metadata:XML RDF-CHPDL DC-XML DC-RDF
:
Copy citation
Share:Bookmark and Share

Secondary language

Language:English
Title:Query by humming on folk song collections
Abstract:
QBH systems are designed to identify the most similar songs in database using hummed query. The process begins by capturing a hummed query on the user side, continues with its transcription using pitch detection algorithms and ends with user being presented the results of search algorithms. The goal of this thesis was to examine existing QBH systems in order to find the most optimal one for use in web application EtnoFletno, which was followed by its implementation. Algorithms YIN and probabilistic YIN were considered and tested for query transcription phase of the target system. Several search algorithms were implemented and tested as well, including DTW, Edit Distance, Spring, SMGT and SMBGT. Transcription and search algorithms had to be optimized for usage in EtnoFletno, hence the testing database contained Slovenian folk songs. Final results show that the transcription algorithm probabilistic YIN was better than its predecessor YIN and algorithm SMBGT outperformed all other search algorithms. It is also shown in the results that algorithm SMBGTs parameters should be used in two different predefined ways considering users singing skills.

Keywords:query by humming, transcription of hummed query, EtnoFletno, Slovenian folk songs, search algorithms, pitch detection algorithms, dynamical programming, probabilistic YIN, subsequence matching

Similar documents

Similar works from RUL:
Similar works from other Slovenian collections:

Back