SPIIRAS Proceedings. 2011. Issue 18. Current version: RUS | EN

Baltiyskiy I.A., Nikolenko S.I.
A PROBABILISTIC GRAPHICAL MODEL FOR THE MUSIC HARMONY SIMILARITY TASK.

Full text: pdf
Informregister ID: 0421100130\0029
URL: http://www.proceedings.spiiras.nw.ru/data/src/2011/18/00/spyproc-2011-18-00-06-en.html
Reference: Baltiyskiy I.A., Nikolenko S.I. A probabilistic graphical model for the music harmony similarity task. // SPIIRAS Proceedings. 2011. Issue 18. Pp. 136-163.
UDC 004.85
Baltiyskiy I.A., Nikolenko S.I. A probabilistic graphical model for the music harmony similarity task..
Abstract. Methods for measuring music similarity allow for implementations of completely automated content-based music recommendation systems (similar to Pandora, but without the manual work of expert musicologists). This paper presents a novel method of measuring music harmony similarity based on an original probabilistic graphical model. The model includes information about the current chord and mode; we introduce a hidden parameter, style, which governs the probability of using of a certain chord within the context of a certain mode, and propose to measure the similarity as a distance between parameter vectors of the probability distribution function for style. Similar to some methods for extracting chord progressions, our model includes neither the rhythmic information nor the dependencies between neighboring chords. We describe the implementation of our model done with the Infer.NET system and show experimental results on generated data. The results of experiments with real-world data are negative, which indicates that simple bag-of-chords models are not suitable for the music similarity task
Keywords: music information retrieval, recommendation systems, harmony, similarity task, graphical models, probabilistic models.
References
  1. Allan H., Mullensiefen D., Wiggins G. Methodological Considerations In Studies Of Musical Similarity // Proceedings of the International Conference on Music Information Retrieval (ISMIR 2007) (Vienna, Austria, September 23–27, 2007) URL: http://ismir2007.ismir.net/proceedings/ISMIR2007_p473_allan.pdf
  2. Anderson C. The Long Tail: Why the Future of Business Is Selling Less of More. New York: Hyperion, 2006. 256 pp.
  3. Bas de Haas W., Veltkamp R., Wiering F. Tonal Pitch Step Distance: a Similarity Measure for Chord Progressions // Proceedings of the Ninth International Conference on Music Information Retrieval (ISMIR 2008) (Philadelphia, Pennsylvania, September 14–18, 2008) Pp. 51–56.
  4. Bishop C. Pattern Recognition and Machine Learning. Springer, 2006. 738 pp.
  5. Cano P., Koppenberger M., Wack N. Content-based music audio recommendation // Proceedingsof the 13th annual ACM international conference on Multimedia (MULTIMEDIA’ 05) (Singapore, November 6–11, 2005). Pp. 211–212.
  6. Castelluccio M. The Music Genome Project // Strategic Finance. 2006. № 88 (6). Pp. 57–58.
  7. Fink, D. A Compendium of Conjugate Priors. URL: http://www.johndcook.com/CompendiumOfConjugatePriors.pdf.
  8. Harte C., Sandler M., Abdalla, S., Gomez E. Symbolic Representation of Musical Chords: A Proposed Syntax for Text Annotations // In J. Reiss & G. Wiggins (Eds.), Proceedings of the Sixth International Conference on Music Information Retrieval (ISMIR 2005) (London, UK, September 11–15, 2005). London, UK: University of London, 2005. Pp. 312–319.
  9. Amazon.com: Online Shopping for Electronics, Apparel, Computers, Books, DVDs & more. URL: http://amazon.com
  10. List of The Beatles songs // Википедия. [2011—2011]. Дата обновления: 11.11.2011. URL: http://en.wikipedia.org/w/index.php?title=List_of_The_Beatles_songs&oldid=460172268
  11. Reference Annotations: The Beatles. URL: http://isophonics.net/content/referenceannotations-beatles
  12. Million Song Dataset. URL: http://labrosa.ee.columbia.edu/millionsong/
  13. Last.fm - Listen to internet radio and the largest music catalogue online. URL: http://last.fm
  14. Pandora Radio — Listen to Free Internet Audio, Find New Music.URL: http://pandora.com
  15. Apple - iTunes - Everything you need to be entertained. URL: http://www.apple.com/itunes/
  16. Secrets of The Beatles // SeeChord. URL: http://www.seechord.co.uk/song-writing/secretsof-the-beatles/
  17. Anderson C. The Long Tail // Журнал Wired: [сайт]. URL: http://www.wired.com/wired/archive/12.10/tail.html
  18. Netflix - Watch TV Shows Online, Watch Movies Online. URL: https://www.netflix.com/
  19. Joyce J. Pandora and the Music Genome Project / Scientific Computing. 2006. № 23 (10). Pp. 14, 40–41.
  20. Krumhansl C. Cognitive Foundations of Musical Pitch. New York: Oxford University Press, 1990. 318 pp.
  21. Krumhansl C., Shepard R. Quantification of the hierarchy of tonal functions within a diatonic context // Journal of Experimental Psychology: Human Perception and Performance. 1979. № 5. Pp. 579–594.
  22. Minka T. Expectation propagation for approximate Bayesian inference // In Breese, Jack S. And Koller, Daphne (eds.), Proceedings of the 17th Conference on Uncertainty in Artificial Intelligence (Seattle, WA, August 2–5, 2001). Massachusetts: Morgan Kaufmann, 2001. Pp. 362–369.
  23. Minka T., Winn J., Guiver J., Knowles D. Infer.NET 2.4, Microsoft Research Cambridge, 2010. URL: http://research.microsoft.com/infernet
  24. Muller M. Information Retrieval for Music and Motion. Berlin: Springer-Verlag, 2007. 334 pp.
  25. Pickens J., Crawford T. Harmonic Models for Polyphonic Music Retrieval // Proceedings of International Conference on Information and Knowledge Management (CIKM 2002) (McLean, VA, November 4–9, 2002). Pp. 430–437.
  26. Pollack A.W. Notes on … Series, 1989–2000 [Электронный ресурс]. The 'Official' rec.music.beatles Home Page: [сайт]. URL: http://www.recmusicbeatles.com.
  27. Scaringella N., Zoia G., Mlynek D. Automatic genre classification of music content: a survey // IEEE Signal Processing Magazine. 2006. Vol. 23, № 2. Pp. 133-141.
  28. The HDF Group. Hierarchical data format version 5, 2000-2010. URL: http://www.hdfgroup.org/HDF5.
  29. Bertin-Mahieux T., Ellis D.P.W., Whitman B., Lamere P. The million song dataset // In Proceedings of the 12th International Society for Music Information Retrieval Conference (ISMIR 2011), (Miami, Florida, October 24–28, 2011). URL: http://ismir2011.ismir.net/papers/OS6-1.pdf
  30. Yoshii K., Goto M., Komatani K., Ogata T., Okuno H.G. Hybrid collaborative and contentbased music recommendation using probabilistic model with latent user preferences // Proceedings of the 7th International Conference on Music Information Retrieval (ISMIR 2006) (Victoria, Canada, October 8–12, 2006). URL: http://ismir2006.ismir.net/PAPERS/ISMIR0647_Paper.pdf.
  31. Абызова Е.А. Гармония: Учебник. М.: Музыка, 2008, 383с.
  32. Балтийский И.А. Сравнение методов вычисления признаков для задачи поиска музыки по голосу. Бакалаврская работа, СПбГУ ИТМО, 2009.
  33. Балтийский И.А., Николенко С.И. Обзор графических вероятностных моделей гармонии для анализа музыкальных произведений. // Труды СПИИРАН. СПб.: Наука, 2011.
  34. Балтийский И.А., Николенко С.И. Системы query-by-humming: обзор подходов и схема платформы для экспериментов // Труды СПИИРАН. СПб.: Наука, 2008. Вып. 7. С. 75–92.
  35. Тулупьев А.Л, Николенко С.И., Сироткин А.В. Байесовские сети доверия: логико- вероятностный вывод в ациклических направленных графах. СПб.: Изд-во С.-Петерб. Ун-та, 2009, 400 с.
  36. Тулупьев А.Л., Николенко С. И., Сироткин А. В. Байесовские сети: логико-вероятностный подход. СПб.: Наука, 2006. 608 с.
  37. Тулупьев А.Л. Алгебраические байесовские сети: реализация логико-вероятностного вывода в комплексе java-программ // Труды СПИИРАН. СПб.: Наука, 2009. Вып. 8. С. 191–232.
  38. Тулупьев А.Л. Задача локального автоматического обучения в алгебраических байесовских сетях: логико-вероятностный подход // Труды СПИИРАН. 2008. Вып. 7. СПб.: Наука, 2008. С. 11–25.
  39. Фильченков А.А., Тулупьев А.Л. Структурный анализ систем минимальных графов смежности // Труды СПИИРАН. 2009. Вып. 11. С. 104–127.

Baltiyskiy Igor Andreevich — M. Sc. in Applied Mathematics and Informatics; student, Saint-Petersburg State University of Fine Mechanics and Optics.

The number of publications — 2.

E-mail: iosank@gmail.com
Address: Sablinskaya street, 14, Saint-Petersburg, 197101, Russia
Office phone: +7(921)792-73-22

Scientific advisor — S.I. Nikolenko.

Nikolenko Sergey Igorevich — Ph.D.; Researcher at the Laboratory of Mathematical Logic, St. Petersburg Academic University.

Research interests: music information retrieval, probabilistic models, music theory, digital signal processing The number of publications — 60.

E-mail: sergey@logic.pdmi.ras.ru
Web: http://logic.pdmi.ras.ru
Address: nab. r. Fontanki, 27, St. Petersburg, 191023, Russia
Office phone: +7(812)3124058
Fax: +7(812)3105377
Full text: pdf
Informregister ID: 0421100130\0029
URL: http://www.proceedings.spiiras.nw.ru/data/src/2011/18/00/spyproc-2011-18-00-06-en.html
Reference: Baltiyskiy I.A., Nikolenko S.I. A probabilistic graphical model for the music harmony similarity task. // SPIIRAS Proceedings. 2011. Issue 18. Pp. 136-163.