Postprocessing for Tonality and Repeatability, and Average Neural Networks for Training Multiple Songs in Automatic Composition

자동작곡에서 조성과 반복구성을 위한 후처리 방법 및 다수 곡 학습을 위한 평균 신경망 방법

  • Kim, Kyunghwan (Department of Electronics and Information Engineering, Hansung University) ;
  • Jung, Sung Hoon (Department of Electronics and Information Engineering, Hansung University)
  • 김경환 (한성대학교 전자정보공학과) ;
  • 정성훈 (한성대학교 전자정보공학과)
  • Received : 2016.11.22
  • Accepted : 2016.12.14
  • Published : 2016.12.25


This paper introduces a postprocessing method, an iteration method for melody, and an average neural network method for learning a large number of songs in order to improve musically insufficient parts in automatic composition using existing artificial neural network. The melody of songs composed by artificial neural networks is produced according to the melodies of trained songs, so it can not be a specific tonality and it is difficult to have a repetitive composition. In order to solve these problems, we propose a postprocessing method that converts the melody composed by artificial neural networks into a melody having a specific tonality according to music theory and an iteration method for melody by iteratively composing measure divisions of artificial neural networks. In addition, the existing training method of many songs has some disadvantages. To solve this problem, we adopt an average neural network that is made by averaging the weights of artificial neural networks trained each song. From some experiments, it was confirmed that the proposed method solves the existing problems.

본 논문에서는 기존의 인공신경망을 이용한 자동작곡에서 음악적으로 부족한 부분을 개선하기 위해 조성을 후처리하는 방법과 멜로디에 반복성을 주는 방법 그리고 다수의 곡을 학습하기 위한 평균 신경망 방법을 제안한다. 인공신경망을 이용하여 작곡된 곡의 멜로디는 인공신경망에 학습된 곡의 멜로디에 따라서 출력되는 것으로 음악적으로 특정한 조성에 맞는 곡이 출력되지 않으며 또한 반복적인 멜로디 구성이 나오기 어렵다. 본 논문에서는 이를 해결하기 위하여 인공신경망이 출력한 멜로디를 음악이론에 따라서 특정한 조성으로 후처리하는 방법과 마디구분을 반복적으로 구성하여 멜로디 진행에 반복을 주는 방법을 제안한다. 또한 기존 연구에서 사용한 다수의 곡을 학습하는 방법은 여러 가지 단점이 있었다. 이를 해결하기 위하여 다수의 곡을 학습하는 방법으로 각 곡을 학습한 인공신경망의 가중치를 평균하여 만든 평균 인공신경망을 사용하는 것을 제안한다. 제안한 방법을 적용하여 작곡한 결과 제안한 방법이 기존의 문제점을 해결하는 것을 확인할 수 있었다.



Supported by : 한성대학교


  1. J. Cho, E. M. Ryu, J. Oh, and S. H. Jung, "Training Method of Artificial Neural Networks for Implementation of Automatic Composition Systems," KIPS Transactions on Software and Data Engineering, vol. 3, no. 8, pp. 315-320, Aug. 2014.
  2. J.-W. Oh, J.-H. Song, K-H.. Kim and S. H. Jung, "Automatic Composition Using Training Capability of Artificial Neural Networks and Chord Progression," Journal of Korea Multimedia Society, vol. 18, no. 11, pp. 1358-1366, Nov. 2015.
  3. B. Johanson and R. Poli, "GP-Music: An Interactive Genetic Programming System for Music Generation with Automated Fitness Raters" Proceedings of the Third Annual Conference, pp. 181-186, 1998.
  4. N. Tokui and H. Iba, "Music Composition with Interactive Evolutionary Computation," Proceedings of the Third International Conference on Generative Art, pp. 215-226, 2000.
  5. A. Santos, B. Arcay, J. Dorado, J. Romero, and J. Rodriguez, "Evolutionary Computation Systems for Musical Composition," Proceedings of the International Conference Acoustic and Music: Theory and Applications, pp. 97-102, 2000.
  6. C. Chen and R. Miikkulainen, "Creating Melodies with Evolving Recurrent Neural Networks," Proceedings of the 2001 International Joint Conference on Neural Networks, pp. 2241-2246, 2001.
  7. Debora C. Correa, Alexandre L. M. Levada, Jose H. Saito, and Joao F. Mari, "Neural network based systems for computer-aided musical composition: supervised x unsupervised learning," Proceeding SAC '08 Proceedings of the 2008 ACM symposium on Applied computing, pp. 1738-1742, 2008.
  8. T. Oliwa and M. Wagner, "Composing Music with Neural Networks and Probabilistic Finite-State Machines," Applications of Evolutionary Computing: EvoWorkshops 2008, pp. 503-508, 2008.
  9. H. Kim, B. Kim, and B. Zhang, "Learning music and generation of crossover music using evolutionary hypernetworks," Proceedings of Korea Computer Congress 2009, pp. 134-138, 2009.
  10. G. Bickerman, S. Bosley, P. Swire, and Rober M. Keller, "Learning to Create Jazz Melodies Using Deep Belief Nets," Proceedings of the International Conference on Computational Creativity, pp. 228-237, 2010.
  11. Andres E. Coca, Roseli A. F. Romero, and Liang Zhao, "Generation of composed musical structures through recurrent neural networks based on chaotic inspiration," Proceedings of International Joint Conference on Neural Networks," pp. 3220-3226, 2011.
  12. J. D. Fernandez and F. Vico, "AI Methods in Algorithmic Composition: A Comprehensive Survey," Journal of Artificial Intelligence Research," vol. 48, pp. 513-582, 2013.