人工智能作曲发展的现状和趋势探究  

Research on the Status Quo and Trend of the Development of Artificial Intelligence Composition

作　　者：周莉[1] 邓阳 ZHOU Li;DENG Yang

机构地区：[1]中国地质大学艺术与传媒学院,湖北武汉430074

出　　处：《艺术探索》2018年第5期107-111,共5页Arts Exploration

基　　金：2016年度教育部人文社会科学研究规划基金项目"扬琴音乐机器人及其智能识谱与演奏的关键技术研究"(16YAZH080)

摘　　要：随着人工智能技术的快速发展,多种基于人工智能的算法被运用到作曲中,主要包括马尔科夫链、神经网络、遗传算法,以及多种混合型算法等。这些作曲算法在当前实践中存在着不同的优劣势,所创作的音乐作品风格和体裁还比较单一,且可听性不高。随着用户对作曲系统智能化程度要求的提高,人工智能作曲的发展将呈现以下趋势:以多算法组合优化为方向,以多源音乐情感识别与优化推理为基础,中国民族音乐人工智能作曲系统发展空间广阔,人工智能作曲与机器人紧密结合。With the rapid development of artificial intelligence technology, varieties of artificial intelligence-based algorithms are applied to the composition, including Markov Chain, Neural Networks, Genetic Algorithm, and a variety of hybrid algorithms. These composing algorithms have different advantages and disadvantages in the current practice. The style and genre of the created works are relatively simple and audible. With the increase of users’ requirements for artificial intelligence composition, the development of artificial intelligence composition will show the following trends： multi-algorithm combination optimization as the direction, based on multi-source music emotion recognition and optimization reasoning, the development space of Chinese national music artificial intelligence composing system is vast, and artificial intelligence composition is closely integrated with the robot.

关 键 词：算法作曲 人工智能作曲 音乐情感计算 混合型算法作曲 

分 类 号：J61[艺术—音乐]