COMPOSITION OF MUSIC WITH DUAL INSTRUMENTS BASED ON MARKOV MODEL

Yıl 2020, , 1236 - 1247, 25.12.2020

Muratcan Kılıç H. İrem Türkmen

https://doi.org/10.21923/jesd.558720

Öz

Music has been implemented as a human-made sequence of tones throughout the history. In the last sixty years, there are ongoing studies about automated composition of music by computers. In this study, we propose a novel system, referred to as EMERSON, which exploits Markov process in order to perform automated composition of music. It has been analyzed that whether automatic composed music is approximated or not to the performance of the human-made music. For this purpose, differently from the works in technical literature, two-instrument pieces are generated by the proposed study. First instrument sequence is trained by using human-made composition while sequence of second instrument is trained by using the produced sequence of the first instrument. Performance evaluation of generated music is performed by surveying. In the Turing test part of survey it has been achieved 43.32% success rate, whereas in liking test a success rate of 5.62 points out of 10 is calculated.

Anahtar Kelimeler

Markov Model, Automated Music Composition, Dual Instruments

MARKOV MODEL TABANLI ÇİFT ENSTRUMANLI MÜZİK BESTELEME

Öz

Müzik, tarih boyunca insan eliyle üretilen tonlar dizilimi olarak gerçeklenmiştir. Son altmış yılda müziğin otomatik olarak bilgisayarlar yardımıyla bestelenmesi ile ilgili çalışmalar yapılmaktadır. Bu çalışmada, müziğin otomatik olarak üretilmesinde Markov süreci yaklaşımının kullanıldığı özgün bir müzik besteleme sistemi, EMERSON, sunulmuştur. EMERSON’ın insan yapımı müziklerin performansına ne kadar yaklaşabileceğinin tahlili yapılmıştır. Bunun için önerilen sistemde literatürdeki diğer çalışmalardan farklı olarak iki enstrümanlı parçalar üretilmiştir. Birinci enstrümanın eğitiminde bir insan yapımı beste, ikinci enstrümanın eğitiminde birinci enstrümanın müzik sekansı kullanılmıştır. Üretilen müziklerin performansı anket yoluyla ölçülmüştür. Anketin Turing testi kısmında %43.32 oranında başarı oranı, beğeni testinde ise 10 üzerinden ortalama 5.62 puan elde edilmiştir.

Anahtar Kelimeler

Markov Model, Otomatik Müzik Besteleme, Çift Enstruman

Kaynakça

• Ames, C. (1987). Automated composition in retrospect: 1956-1986. Leonardo, 20 (2), 169–185.
• Ariza, C. (2011). Two pioneering projects from the early history of computer-aided algorithmic composition. Computer Music Journal, 35 (3), 40-56.
• Boenn, G., Brain, M., De Vos, M., & Ffitch, J. (2011). Automatic music composition using answer set programming. Theory and practice of logic programming, 11 (2-3), 397–427.
• Burton, A. R., & Vladimirova, T. (1999). Generation of musical sequences with genetic techniques. Computer Music Journal, 23 (4), 59–73.
• Chen, C.-C., & Miikkulainen, R. (2001). Creating me-lodies with evolving recurrent neural networks. In Neural networks, 2001. proceedings. ijcnn’01. international joint conference on (Vol. 3, pp. 2241–2246)
• Colombo, F., Muscinelli, S. P., Seeholzer, A., Brea, J.,& Gerstner, W. (2016, June). Algorithmic Composition of Melodies with Deep Recurrent Neural Networks. ArXiv e-prints.
• de la Puente, A. O., Alfonso, R. S., & Moreno, M. A. (2002). Automatic composition of music by means of grammatical evolution. In Acm sigapl apl quote quad (Vol. 32, pp. 148–155).
• Frequencies of musical notes, a4 = 440 hz.(n.d.). Retrieved from https://pages.mtu.edu/ suits/notefreqs.html
• Garcia Salas, H. A., Gelbukh, A., Calvo, H., & Galindo Soria, F. (2011, 12). Automatic Music Composition with Simple Probabilistic Generative Grammars. Polibits, 59 - 65.
• HEDGES, S. A. (1978). Dice music in the eighteenth century. Music and Letters, 59 (2), 180-187. doi: 10.1093/ml/59.2.180
• Herremans, D., Weisser, S., Sörensen, K., & Conklin,D. (2015). Generating structured music for bagana using quality metrics based on markov models. Expert Systems with Applications, 42 (21), 7424 – 7435. doi: https://doi.org/10.1016/j.eswa.2015.05.043
• Hiller, L. A., & Baker, R. A. (1964). Computer cantata: A study in compositional method. Perspectives of New Music, 3 (1), 62–90.
• Holtzman, S. R. (1981). Using generative grammars for music composition. Computer Music Journal, 5 (1), 51–64
• Hornel, D., & Menzel, W. (1998). Learning musical structure and style with neural networks. Computer Music Journal, 22 (4), 44–62.
• Lo, M., & Lucas, S. M.(2006). Evolving musical sequences with n-gram based trainable fitness functions. In 2006 ieee international conference on evolutionary computation (p. 601-608). doi: 10.1109/CEC.2006.1688365
• Long, P. (1977). Relationships between Pitch Memory in Short Melodies and Selected Factors. Journal of Research in Music Education, 25(4), 272-282. Retrieved from http://www.jstor.org/stable/3345268
• Mazurowski, L. (2012, Sept). Computer models for algorithmic music composition. In 2012 federated conference on computer science and information systems (fedcsis) (p. 733-737).
• Merwe, A. V. D., & Schulze, W. (2011, March). Music generation with markov models. IEEE MultiMedia, 18 (3), 78-85. doi: 10.1109/MMUL.2010.44
• Muñoz, E., Cadenas, J. M., Ong, Y. S., & Acampora, G. (2016, Feb). Memetic music composition. IEEE Transactions on Evolutionary Computation, 20 (1), 1-15. doi: 10.1109/TEVC.2014.2366871
• Papadopoulos, A., Roy, P., & Pachet, F. (2014). Avoiding plagiarism in markov sequence generation. In Proceedings of the twenty-eighth aaai conference on artificial intelligence (pp. 2731–2737). AAAI Press.
• Roig, C., Tardón, L. J., Barbancho, I., & Barbancho, A. M. (2014). Automatic melody composition based on a probabilistic model of music style and harmonic rules. Knowledge-Based Systems, 71, 419 – 434
• Roig, C., Tardón, L. J., Barbancho, I., & Barbancho, A. M. (2018). A non-homogeneous beat-based harmony markov model. Knowledge-Based Systems, 142, 85 - 94. doi: https://doi.org/10.1016/j.knosys.2017.11.027
• Sanford, L., (2007). C# MIDI Toolkit https://www.codeproject.com/Articles/6228/C-MIDI-Toolkit
• Tanaka, T., Nishimoto, T., Ono, N., & Sagayama, S. (2010). Automatic music composition based on counterpoint and imitation using stochastic models.
• Toivanen, J. M., Toivonen, H., & Valitutti, R. (n.d.). Automatical composition of lyrical songs.
• Westergaard, P., & Hiller, L. A. (1959). Journal of Music Theory, 3 (2), 302–306.
• Zhao, Y., Liu, L., Huang, Y., & Fang, N. (2019). Automatic Composition of Music by LSTM. In 2018 International Workshop on Education Reform and Social Sciences. Atlantis Press.