2010
DOI: 10.1016/j.sigpro.2009.06.017
|View full text |Cite
|
Sign up to set email alerts
|

Pitch-frequency histogram-based music information retrieval for Turkish music

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

1
43
0
10

Year Published

2014
2014
2024
2024

Publication Types

Select...
4
4

Relationship

1
7

Authors

Journals

citations
Cited by 49 publications
(54 citation statements)
references
References 28 publications
1
43
0
10
Order By: Relevance
“…Further investigations revealed that the assignment rules based on these membership functions failed to assign any phrase to these makamlar, preventing the computation of a precision rate as well as a subsequent F-measure. Similar observations were reported by previous makam classification studies (such as Ünal, Bozkurt, and Karaosmanoglu 2014;Gedik and Bozkurt 2010) since the pitch class distributions of Beyati and Uşşak makamlar are very similar (as can be seen in Figure 6), and from music theory we know that Rast and Hüseyni pieces may include a large amount of Uşşak makam phrases. It is clear that new features are needed to capture the characteristics that differentiate specifically these two makamlar.…”
Section: Resultssupporting
confidence: 88%
See 1 more Smart Citation
“…Further investigations revealed that the assignment rules based on these membership functions failed to assign any phrase to these makamlar, preventing the computation of a precision rate as well as a subsequent F-measure. Similar observations were reported by previous makam classification studies (such as Ünal, Bozkurt, and Karaosmanoglu 2014;Gedik and Bozkurt 2010) since the pitch class distributions of Beyati and Uşşak makamlar are very similar (as can be seen in Figure 6), and from music theory we know that Rast and Hüseyni pieces may include a large amount of Uşşak makam phrases. It is clear that new features are needed to capture the characteristics that differentiate specifically these two makamlar.…”
Section: Resultssupporting
confidence: 88%
“…Ünal, Bozkurt, and Karaosmanoglu (2014) have studied the automatic makam classification problem using n-grams where they report that even 1-gram information, which corresponds to simply pitch class distributions, is highly discriminative. Gedik and Bozkurt (2010) and Ioannidis, Gómez, and Herrera (2011) have shown that pitch distributions can be effectively used for makam classification of audio data.…”
Section: Membership Functionmentioning
confidence: 99%
“…This difference is equal to one Turkish kuma (or cumma). Each octave consists of 53 logarithmically equal kuma, and since the octave consists of 1200 logarithmically equal cents, the one kuma is about 22.6 cents [35]. So accordingly, the lowerstep interval from Ed in rast and huzam is 7 kuma, and the upper step is 6, in bayat it is just the opposite.…”
Section: Theoretical Background Of the Modelmentioning
confidence: 97%
“…Kullanılan [35] Makam tanıma N-gram modeli MusicXML Gedik ve Bozkurt [32] Makam tanıma Ses frekans histogramı Ses dosyaları Darabi vd. [25] Dastgah ve makam tanıma Örüntü tanıma Ses dosyaları Abdoli [26] Dastgah sınıflandırma Bulanık mantık Ses dosyaları Şentürk vd.…”
Section: Nota Dönüşümü (Note Transformation)unclassified
“…Klasik Türk Müziği eserlerinin makamlarına göre sınıflandırılması konusunda literatürdeki ilk çalışmayı yapan Gedik ve Bozkurt [31], sonraki çalışmalarında, ses frekans histogramları kullanılarak otomatik ton tespiti ile makam tanıma yapmıştır [32]. Bu çalışmada yazarlar, ses frekans histogramlarının Batı Müziği eserlerine başarı ile uygulanan bir teknik olmasına rağmen, Türk Müziği eserlerinde histogramların yanı sıra makam tahmininde önemli bir yere sahip nota dizilişleri ve kalıplar bulunduğundan bahsetmektedirler.…”
Section: Gi̇ri̇ş (Introduction)unclassified