location:Home > 2021 Vol.4 Jun.No.2 > Automatic Generation Method of Piano Melody Based on Improved Neural Network

2021 Vol.4 Jun.No.2

  • Title: Automatic Generation Method of Piano Melody Based on Improved Neural Network
  • Name: Xin Wang
  • Company: Zhengzhou University School of Music
  • Abstract:

    There are defects in the music coherence time structure of automatic melody generation of piano music, and there is a dependency between manual input tracks, which affects the quality of music generation. Based on improved neural network, an automatic generation method of piano melody is proposed. The main melody of piano music is extracted, and the specific notes are obtained according to the mapping relationship between the fundamental frequency and the keys. The feature vector of melody notes of piano music is constructed to make the note sequence better describe the note relationship. The self attention mechanism is introduced into the neural network to reduce the time dependence of the input track. According to the input note characteristics, an automatic melody generation model is established based on improved neural network. The experimental results show that the note span and qualified note proportion of the automatic melody generation method of piano music based on improved neural network are higher than those based on Markov model and LSTM, which can improve the quality of piano music.

     

  • Keyword: improved neural network; automatic generation; music generation; melody generation; piano music; music melody;
  • DOI: 10.12250/jpciams2021090222
  • Citation form: Xin Wang.Automatic Generation Method of Piano Melody Based on Improved Neural Network[J]. Computer Informatization and Mechanical System,2021,Vol.4,pp.13-19.
Reference:


References

[1] WANG Tao, JIN Cong, LI Xiaobing, et al. Multi-track music generative adversarial network based on Transformer[J]. Journal of Computer Applications, 2021, 41(12):3585-3589.

[2] Bai Yong, Qi Lin, Tie Yun. MUSIC GENERATION BASED ON REINFORCEMENT LEARNING ACTOR-CRITIC[J]. Computer Applications and Software, 2020, 37(5):118-122,182.

[3] GUO Linghua, LIU Minglei, DING Tingwen, et al. Research on algorithms for music melody generation from digital image based on LabVIEW[J]. Journal of Xi'an University of Technology, 2020, 36(1):65-71.

[4] Fan Chunlong, Zhang Zhenxin, Ding Sanjun, et al. Interactive melody generation based on variational autoencoder[J]. Application Research of Computers, 2021, 38(2):479-483,488.

[5] Kp A ,  Af B ,  Mnv C . Artificial intelligence methods for music generation: a review and future perspectives - ScienceDirect[J]. Nature-Inspired Computation and Swarm Intelligence, 2020:217-245.

[6] Krishnan V P ,  Rajarajeswari S ,  V  Krishnamohan, et al. Music Generation Using Deep Learning Techniques[J]. Journal of Computational and Theoretical Nanoscience, 2020, 17(9):3983-3987.

[7] Duarte A . Algorithmic interactive music generation in videogames[J]. SoundEffects - An Interdisciplinary Journal of Sound and Sound Experience, 2020, 9(1):38-59.

[8] Li G ,  Ding S ,  Li Y , et al. Music generation and human voice conversion based on LSTM[J]. MATEC Web of Conferences, 2021, 336(3):06015.

[9] Jin C ,  Wang T ,  Liu S , et al. A Transformer-Based Model for Multi-Track Music Generation[J]. International Journal of Multimedia Data Engineering and Management, 2020, 11(3):36-54.

[10] Basha S ,  Vinakota S K ,  Pulabaigari V , et al. AutoTune: Automatically Tuning Convolutional Neural Networks for Improved Transfer Learning[J]. Neural Networks, 2021, 133:112-122.

[11] LIN Zhe, QUAN Hai-yan. BP Neural Network Learning Algorithm Using Surface-Simplex Swarm Evolution[J]. Computer Simulation, 2020, 37(3):270-274.

[12] ZHANG Chi, GUO Yuan, LI Ming. Review of Development and Application of Artificial Neural Network Models[J]. Computer Engineering and Applications, 2021, 57(11):57-69.

[13] He Jian-qiang. Research and Application of Model Optimization Based on Artificial Neural Network[J]. Electronics Quality, 2020, (2):32-35.

[14] Herzog S , Tetzlaff C , Wrgtter F . Evolving artificial neural networks with feedback[J]. Neural Networks, 2020, 123:153-162.

[15] Wu J ,  Hu C ,  Wang Y , et al. A Hierarchical Recurrent Neural Network for Symbolic Melody Generation[J]. IEEE Transactions on Cybernetics, 2020, 50(6):2749-2757.


 

Tsuruta Institute of Medical Information Technology
Address:[502,5-47-6], Tsuyama, Tsukuba, Saitama, Japan TEL:008148-28809 fax:008148-28808 Japan,Email:jpciams@hotmail.com,2019-09-16