| Autor: |
Siphocly, Nermin Naguib J., El-Horbaty, El-Sayed M., Salem, Abdel-Badeeh M. |
| Předmět: |
|
| Zdroj: |
Egyptian Computer Science Journal; Sep2019, Vol. 43 Issue 3, p49-64, 16p |
| Abstrakt: |
Computer music generation systems aid musicians in creating and producing high quality music using computers. In the few past years, various intelligent techniques have been used to teach computers ways of generating music either through composition, improvisation, or expressive music playing. This paper presents a comprehensive analysis of the recent publications in music generation field. We particularly focus on music composition applications, especially those adopting algorithmic composition techniques. Our analysis covers the most recently adopted techniques such as deep neural networks and generative adversarial networks. The importance of our analysis is to give insight to researchers of this area about the most suitable techniques to choose from to perform each musical task. Our study shows that the most successful commonly used technique for improvisation is factor oracles. As for music expressiveness which is a rather new field, mathematical models and neural networks achieved the best results. Composition systems have various branches; however, the state-of-the-art technique for composition applications is the generative adversarial networks. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
