Emotion Recognition of Western Tonal Music Using Support Vector Machine  

作　　者：WANG Muyuan ZHANG Naiyao ZHU Hancheng 

机构地区：[1]Department of Automation, Tsinghua University, Beijing 100084, China [2]Art Education Center, Tsinghua University, Beijing 100084, China

出　　处：《Chinese Journal of Electronics》2006年第1期74-78,共5页电子学报（英文版）

基　　金：This work is supported by the National Natural Science Foundation of China（No.60174015） and the Basic Research Foundation of Tsinghua University.

摘　　要：The automatic emotion recognition of music is helpful for music indexing, retrieval, and understanding. In this paper, we present a scheme to recognize the emotional meaning of western tonal music. The recognition process consists of four steps： first, a 2-dimensional‘Emotion Plane＇ model is constructed as music emotion＇s taxonomy. Second, novel musical features are extracted from MIDI files, according to psychological and music perception theories in western cultures. Third, the Support vector machine （SVM） is trained using the extracted features. Finally, the trained support vector machine can be used to classify the music emotions into six classes, according to the emotion plane model. Satisfying experimental results are obtained on western tonal music, which indicate the effectiveness of our approach.

关 键 词：Computer music Music emotion recognition Emotion model Western tonal music Support vector machine 

分 类 号：TN912.3[电子电信—通信与信息系统]