Violin Etude Composing based on LSTM Model

Xuanhe Zhang

doi:10.54097/hset.v39i.6493

Authors

Xuanhe Zhang

DOI:

https://doi.org/10.54097/hset.v39i.6493

Keywords:

Machine Learning; Music Composing; Computer Science; Deep Learning.

Abstract

After the development of deep learning and computer music, there is more and more electronic music composed by computers and algorithms. However, there are very few projects for violin etudes created based on artificial intelligence. In 2020, The Shanghai Symphony Orchestra Concert Hall hosted a special concert of Mozart's 80-year-old works, the last three symphonic pieces were generated by the AI based on learning a large amount of Mozart's past music, which inspire me. In this paper, classical violin etude (i.e., Rudolphe Kreutzer 42 Studies) is used for the data. Based on several studies, the LSTM model is the framework of this study. In the multiple times of changing seeds and temperature, the creation demo has creativity and similarity, but with less education. The educational meaning is also the future research to focus on. This study makes a step in making violin etudes from the classical violin etudes, which can create new etudes for future violin players. Moreover, the new etudes shed light on Inheriting and innovating the predecessors' composing style and thinking.

Downloads

Download data is not yet available.

References

Cheng Qiyun, Sun Caixin, and Zhang Xiaoxing. Short-Term load forecasting model and method for power system based on complementation of neural network and fuzzy logic. Transactions of China Electrotechnical Society 19.10 (2004): 53-58.

Fang Fang. Research on power load forecasting based on improved BP neural network. Harbin Institute of technology, 2011.

Roads Curtis. The computer music tutorial. MIT press, 1996.

Sherstinsky Alex. Fundamentals of recurrent neural network (RNN) and long short-term memory (LSTM) network. Physica D: Nonlinear Phenomena 404 (2020): 132306.

Selvin Sreelekshmy, et al. Stock price prediction using LSTM, RNN and CNN-sliding window model. 2017 international conference on advances in computing, communications and informatics (icacci). IEEE, 2017.

Yu Yong, et al. A review of recurrent neural networks: LSTM cells and network architectures. Neural computation 31.7 (2019): 1235-1270.

Eck Douglas, and Juergen Schmidhuber. Finding temporal structure in music: Blues improvisation with LSTM recurrent networks. Proceedings of the 12th IEEE workshop on neural networks for signal processing. IEEE, 2002.

Eck Douglas, and Jürgen Schmidhuber. Learning the long-term structure of the blues. International Conference on Artificial Neural Networks. Springer, Berlin, Heidelberg, 2002.

Klapuri Anssi, and Manuel Davy, eds. Signal processing methods for music transcription, 2007.

De Mantaras R. Making music with AI: Some examples. Proceeding of the 2006 conference on Rob Milne: A Tribute to a Pioneering AI Scientist, Entrepreneur and Mountaineer. 2006.

Demo online: retrieved from: https://pan.baidu.com/s/1WNQ-VKvr227UnZurjNfhDA.