情緒導向之自動作曲演算法

The Algorithm of Emotion-based Robot Composition

粘為博(Wei-Po Nien)

指導教授 ： 成維華 ； 黃志方

國立交通大學/工學院/機械工程系所/博士(2015年)

https://doi.org/10.6842/NCTU.2015.00889

摘要

情緒為高層次的心理感受，較能直接反應人的感覺。以情緒為導向的演算法作曲能將情緒的反應快速的轉換至音樂參數，演算法作曲將創作出符合該情緒之音樂。情緒為基礎之演算法不但能符合使用者的感受，亦能讓音樂更為豐富。
演算法作曲音樂平台(ACMP)能讓使用者透過雲端服務取得所需要之音樂，以情緒為基礎之平台將轉換情緒座標為音樂參數，提供系統作為演算法創作的材料。此外情緒座標亦可傳送給機器人，讓機器人根據不同的情緒改變不同類型動作。本研究透過感性工學作為產出音樂的量化方法，由問卷結果可知選擇某情緒所產生之音樂與感性工學詞彙有高度相關。使用者能選擇情緒或感性工學詞彙產生對應之音樂，如此一來能更方便創作屬於自己感受之音樂。

關鍵字

電腦音樂 ； 演算法作曲 ； 自動作曲 ； 情緒音樂 ； 雲端音樂平台

並列摘要

Emotion is a high-level feeling of psychological; it can directly reflect the feeling of people. Emotion-based algorithmic composition converts promptly emotion reactions into music parameters. The algorithmic composition is applied to compose the music pieces corresponding to the emotion of the user. Emotion-based algorithmic composition can not only catch up the user’s feeling but also make the music attach to that feeling.
The algorithmic composition musical platform (ACMP), developed in this dissertation, allows users to acquire their music needs in the cloud service and yields proper music pieces via different user requests. The advanced version, emotion-based platform, can further convert emotion coordinate into music parameters, and create materials to the algorithmic composition. In addition, the emotion coordinate can be delivered to music robots which will do different actions depending on different emotions of the users. Kansei engineering is adopted as assessment method for yielding music, and the result of questionnaire shows that the music generated by the system correlate significantly Kansei engineering adjective. User can select emotion or Kansei engineering adjective to yield proper music, so that user can create the music with their perception.

並列關鍵字

Computer music ； Algorithmic composition ； Automated composition ； Emotional music ； ACMP

參考文獻

[5] D. Cope, Virtual music: computer synthesis of musical style: MIT press, 2004.

[6] D. Cope, Computers and musical style: AR Editions, Inc., 1991.

[7] L. A. Hiller and L. M. Isaacson, Experimental Music; Composition with an electronic computer: Greenwood Publishing Group Inc., 1979.

[8] H. M. Lin and C. F. Huang, "An Algorithmic Composition Method Of Monophonic Pitches Based On The Frequency Relationship Of Melody Motion," the International Computer Music Conference, New York, USA, 2010.

[9] L. Rabiner and B.-H. Juang, "An introduction to hidden Markov models," ASSP Magazine, IEEE, vol. 3, pp. 4-16, 1986.