算法作曲(algorithmic composition),是通过某种形式化的过程,使作曲者在利用计算机进行音乐创作时的介入程度达到最小的研究。遗传算法的快速发展和应用引起了算法作曲研究者的注意,其高鲁棒性、良好的收敛性以及较快的运算速度等显著的特点启发研究者们可以将其与乐曲创作相结合。音乐作品离不开创作者的“灵感”,一首好的音乐作品一定充满着创作者的感情色彩,而交互式遗传算法作曲可以最大程度的发挥作曲者的主观感情色彩。论文重点介绍了基于交互式遗传算法作曲系统的架构与实现,在编码机制、人工评估及人机交互操作等诸多方面都有创新的设计应用和实现方法。
Algorithmic composition is a kind of research to minimize the degree of intervention of the composer in use of comput- er in music creation by a formal process. Rapid development and application of GA has attracted the attention of researehers in algorithmic music composition, the strengths like high robustness, good convergence and faster operation speed inspire the re- searchers to make it combined with music creation. Music works can't live without "inspiration", a good song must be full of emotion of the creator, and IGA composition can maximize the direct feelings of the composers. The paper mainly introduce the architecture and implementation of the composing system based on IGA, makes innovation and implementation in encoding mechanism, artificial assessment and human-computer interaction and so on.