Automatic music transcription
dc.contributor.advisor | Solak, Ercan | en_US |
dc.contributor.author | Paşmakoğlu, Berk Ekim | en_US |
dc.contributor.other | Işık Üniversitesi, Fen Bilimleri Enstitüsü, Bilgisayar Mühendisliği Yüksek Lisans Programı | en_US |
dc.date.accessioned | 2016-06-09T06:16:09Z | |
dc.date.available | 2016-06-09T06:16:09Z | |
dc.date.issued | 2010 | |
dc.department | Işık Üniversitesi, Fen Bilimleri Enstitüsü, Bilgisayar Mühendisliği Yüksek Lisans Programı | en_US |
dc.description | Text in English ; Abstract: English and Turkish | en_US |
dc.description | Includes bibliographical references (leaves 97-99) | en_US |
dc.description | xvii, 105 leaves | en_US |
dc.description.abstract | Computational music research is spread out of the world in many fields. One of these fields is automatic music transcription. During this thesis, we concentrated on the detection of music notes inside an audio signal. We decided to work on a percussive instrument i.e. piano because percussive onset can be relatively more easier to detect than other types of onset. We benefitted from the signal processing techniques like FFT, low-pass filtering and the statistical methods like Hinkley's CUSUM algorithm and linear regression. We proposed a transcription algorithm applied to a synthetically created audio data which was formed by the notes of middle octave and first five note value types. The algorithm transcribes the music scores with an average accuracy of 96,7 using the tuned parameters. | en_US |
dc.description.abstract | Bilişimsel müzik araştırması bir çok alanda dünyaya yayılmıştır. Bu alanlardan biri de özdevimli müzik çevriyazımıdır. Bu tez sırasında, bir ses iminin içerisindeki müzik notalarının algılanması üzerine yoğunlaştık. Bir vurmalı müzik aleti olan piano üzerine çalışmaya karar verdik çünkü vurmalı nota başlangıçlarının algılanılması diğer nota başlangıç tiplerine göre göreceli olarak daha kolaydır. Hızlı Fourier Dönüşüm'ü ve alçak geçirgen süzgeci gibi im isleme tekniklerinden ve Hinkley'in CUSUM algoritması ve dogrusal regresyon gibi sayımlama yöntemlerinden faydalandık. Orta oktav notalarından ve ilk beş nota değer türlerinden oluşan bireşimsel olarak yaratılmış bir ses verisine uygulanan bir algoritma teklif ettik. Algoritma müzik parçalarını ayarlanmış değiştirgeler kullanarak ortalama yüzde 96,7 bir doğrulukla yazılı biçime dönüştürmektedir. | en_US |
dc.description.tableofcontents | Musical Background | en_US |
dc.description.tableofcontents | What is music? | en_US |
dc.description.tableofcontents | Sound | en_US |
dc.description.tableofcontents | Frequency, Period and Amplitude | en_US |
dc.description.tableofcontents | Musical Note and Pitch | en_US |
dc.description.tableofcontents | Pitch | en_US |
dc.description.tableofcontents | Musical Note | en_US |
dc.description.tableofcontents | Note Value Types | en_US |
dc.description.tableofcontents | Intervals | en_US |
dc.description.tableofcontents | Scales | en_US |
dc.description.tableofcontents | Rhythm of the Music | en_US |
dc.description.tableofcontents | Time Signature | en_US |
dc.description.tableofcontents | Tempo | en_US |
dc.description.tableofcontents | Organization of Thesis | en_US |
dc.description.tableofcontents | Computer Music and MIDI | en_US |
dc.description.tableofcontents | MIDI File Format | en_US |
dc.description.tableofcontents | Header Chunk | en_US |
dc.description.tableofcontents | Track Chunk | en_US |
dc.description.tableofcontents | Variable Length Reading and Writing | en_US |
dc.description.tableofcontents | Variable Length Writing | en_US |
dc.description.tableofcontents | Variable Length Reading | en_US |
dc.description.tableofcontents | The Events in MIDI | en_US |
dc.description.tableofcontents | The Track Chunk MIDI Events | en_US |
dc.description.tableofcontents | The Meta Events | en_US |
dc.description.tableofcontents | Automatic Music Transcription | en_US |
dc.description.tableofcontents | Getting Input Wave File | en_US |
dc.description.tableofcontents | Signal Information Retrieval | en_US |
dc.description.tableofcontents | Getting ‘Times’ | en_US |
dc.description.tableofcontents | Getting ‘Notes’ and ‘Octaves’ | en_US |
dc.description.tableofcontents | Making Single Channel Signal | en_US |
dc.description.tableofcontents | Construction of Amplitude Envelope of Signal | en_US |
dc.description.tableofcontents | Smoothing the Envelope Shape | en_US |
dc.description.tableofcontents | Slope Detector | en_US |
dc.description.tableofcontents | The Model 'Rise-Time' | en_US |
dc.description.tableofcontents | Progressing the Model | en_US |
dc.description.tableofcontents | CUSUM Algorithm | en_US |
dc.description.tableofcontents | Note Segmentation | en_US |
dc.description.tableofcontents | Spurious Attack Elimination | en_US |
dc.description.tableofcontents | Pitch Detection between Successive Note Onsets | en_US |
dc.description.tableofcontents | Note Durations Calculation | en_US |
dc.description.tableofcontents | Note Value Types Detection | en_US |
dc.description.tableofcontents | Note Labels Assignment | en_US |
dc.description.tableofcontents | Proposed Transcription Algorithm | en_US |
dc.description.tableofcontents | Results | en_US |
dc.description.tableofcontents | Data | en_US |
dc.description.tableofcontents | Envelope | en_US |
dc.description.tableofcontents | Smoothing | en_US |
dc.description.tableofcontents | Slope Detection | en_US |
dc.description.tableofcontents | ‘Rise-Time' Model | en_US |
dc.description.tableofcontents | CUSUM Algorithm | en_US |
dc.description.tableofcontents | Note Segmentation | en_US |
dc.description.tableofcontents | Spurious Attack Elimination | en_US |
dc.description.tableofcontents | Pitch Detection | en_US |
dc.description.tableofcontents | Calculation of Note Durations | en_US |
dc.description.tableofcontents | Note Value Types Detection | en_US |
dc.description.tableofcontents | Note Labels Assignment | en_US |
dc.description.tableofcontents | Performance Measurement | en_US |
dc.description.tableofcontents | Experimental Determination of Parameters | en_US |
dc.description.tableofcontents | Variable Length Writing and Reading | en_US |
dc.description.tableofcontents | Pseudocode of Variable Length Writing | en_US |
dc.description.tableofcontents | Pseudocode of Variable Length Reading | en_US |
dc.description.tableofcontents | Note Value Types Detection | en_US |
dc.description.tableofcontents | Pseudocode of Round#1: Finding Centroids | en_US |
dc.description.tableofcontents | Pseudocode of Round#2: Mapping Note Durations | en_US |
dc.description.tableofcontents | Pseudocode of Note Label Assignment | en_US |
dc.identifier.citation | Paşmakoğlu, B. E. (2010). Automatic music transcription. İstanbul: Işık Üniversitesi Fen Bilimleri Enstitüsü. | en_US |
dc.identifier.uri | https://hdl.handle.net/11729/972 | |
dc.institutionauthor | Paşmakoğlu, Berk Ekim | en_US |
dc.language.iso | en | en_US |
dc.publisher | Işık Üniversitesi | en_US |
dc.relation.publicationcategory | Tez | en_US |
dc.rights | info:eu-repo/semantics/openAccess | en_US |
dc.rights | Attribution-NonCommercial-NoDerivs 3.0 United States | * |
dc.rights.uri | http://creativecommons.org/licenses/by-nc-nd/3.0/us/ | * |
dc.subject.lcc | TK5105.8863 .P37 2010 | |
dc.subject.lcsh | Sound -- Recording and reproducing -- Digital techniques. | en_US |
dc.subject.lcsh | MUSIC (Computer system) | en_US |
dc.subject.lcsh | Electronic digital computers. | en_US |
dc.title | Automatic music transcription | en_US |
dc.title.alternative | Özdevimli müzik çeviriyazımı | en_US |
dc.type | Master Thesis | en_US |
dspace.entity.type | Publication |