Open AccessA transformer generative adversarial network for multi‐track music generation
Author(s) -
Jin Cong,
Wang Tao,
Li Xiaobing,
Tie Chu Jie Jiessie,
Tie Yun,
Liu Shan,
Yan Ming,
Li Yongzhi,
Wang Junxian,
Huang Shenze
Publication year - 2022
Publication title -
caai transactions on intelligence technology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.613
H-Index - 15
ISSN - 2468-2322
DOI - 10.1049/cit2.12065
Subject(s) - transformer , generative adversarial network , computer science , artificial neural network , artificial intelligence , speech recognition , deep learning , engineering , electrical engineering , voltage
This study proposes a new generation network based on transformers and guided by the music theory to produce high‐quality music work. In this study, the decoding block of the transformer is used to learn the internal information of single‐track music, and cross‐track transformers are used to learn the information amongst the tracks of different musical instruments. A reward network based on the music theory is proposed, which optimizes the global and local loss objective functions while training and discriminating the network so that the reward network can provide a reliable adjustment method for the generation of the network. The method of combining the reward network and cross entropy loss is used to guide the training of the generator and produce high‐quality music work. Compared with other multi‐track music generation models, the experimental results verify the validity of the model.