IEICE Information and Communication Technology Forum


Session Number:ICTF_5



Intelligent System Based on Deep Learning Technique for Accompaniment Music Generation

Nermin Naguib J. Siphocly,  El-Sayed M. El-Horbaty,  Abdel-Badeeh M. Salem,  


Publication Date:2021/03/24

Online ISSN:2188-5079


PDF download (360.9KB)

Utilizing modern deep learning techniques in music generation is currently a hot research topic. The objective of this paper is to develop an intelligent accompaniment music generator using pix2pix GAN (generative adversarial network) deep networks. We present a novel representation of musical data into images through color encodings. We compare the effect of our suggested data representation technique on the pix2pix network learning as opposed to the traditional binary encoding scheme used in the earlier literature. Our experimental results show that our music representation achieved better results on pix2pix GANs over the traditional representations; reaching a loss function value of 0.001.