第五章 實驗
6.2 未來研究
除了以 Motivic treatment 的方法來探勘動機外，我們可以用其他的方法來探勘動機。
LBDM 的技術可以將音樂分解成許多的音型(Figure)，動機可以由一個以上的音型所構成。
所以我們可以先利用 LBDM 的技術將粗略分段的結果分解成很多個音型。然後利用 Clustering 的技術將這些音型做分群，同一個群組內的音型即可能與同一個動機相關。我們 在分群的時候，計算音型之間的距離，也可以利用動機變化的六種情形。
除了修改傳統探勘重複續列的方法以符合動機特性外，我們可以進一步訂立出符合動 機特性而且與一般重複序列特性不同的的動機樣式(Motive pattern)，以及探勘的方法，以改 進效率或效果。
直 接 利 用 探 勘 出 來 的 動 機 ， 我 們 也 可 以 自 動 產 生 出 音 樂 的 摘 要(Summary 或 Thumbnailing)。因為動機代表音樂的主題，摘要必須呈現音樂中最主要的部份。所以透過 產生出來的動機，我們可以結合成音樂的摘要。
此外，動機也可以作為以音樂內容查詢(Content based music retrieval)時，比對的資料。
相對於對整首音樂作查詢比對，音樂的動機可以改進比對時的效率。
參考文獻
[1] M. A. Bartsch and G. H. Wakefield, “To Catch a Chorus: Using Chroma-Based Representations for Audio Thumbnailing,” Proc. of Workshop on Applications of Signal
Processing to Audio and Acoustics, WASPAA’01, 2001.
[2] R. Bod, “A Memory-Based Model for Music Analysis: Challenging the Gestalt Principles,”
Journal of New Music Research, Vol. 31, pp. 27-37, 2001.
[3] E. Cambouropoulos, “A Formal Theory for the Discovery of Local Boundaries in a Melodic Surface,” Proc. of the III Journees d' Informatique Musicale, 1996.
[4] E. Cambouropoulos, “The Local Boundary Detection Model (LBDM) and its Application in the Study of Expressive Timing,” Proc. of the International Computer Music Conference,
ICMC’01, 2001.
[5] W. Chai and B. Vercoe, “Folk Music Classification Using Hidden Markov Models,” Proc.
of the International Conference on Artificial Intelligence, ICAI’01, 2001.
[6] W. Chai and B. Vercoe, “Music Thumbnailing via Structural Analysis,” Proc. of ACM
International Conference on Multimedia, MM’03, 2003.
[7] W. Chai and B. Vercoe, “Structural Analysis of Musical Signals for Indexing and Thumbnailing,” Proc. of ACM/IEEE Joint Conference on Digital Libraries, JCDL’03, 2003.
[8] Y. J. Chen, “A Fast Repeating Pattern Finding Algorithm for Music Data: A Human Perceptive Approach,” Master Thesis, Department of Electrical Engineering, National Cheng Kung University, Taiwan, 2004.
[9] D. Conklin and C. Anagnostopoulou, “Representation and Discovery of Multiple Viewpoint Patterns,” Proc. of the International Computer Music Conference, ICMC’01, 2001.
[10] R. B. Dannenberg, “A Brief Survey of Music Representation Issues, Techniques and Systems,” Computer Music Journal, Vol. 17, No. 3, pp. 20-30, 1993.
[11] R. B. Dannenberg, B. Thom, and D.Watson, “A Machine Learning Approach to Musical Style Recognition,” Proc. of International Computer Music Conference, ICMC’97, 1997.
[12] C. David, Virtual Music : Computer Synthesis of Musical Style, MIT Press, 2001.
[13] S. Dixon, E. Pampalk, and G. Widmer, “Classification of Dance Music by Periodicity Patterns,” Proc. of International Symposium on Music Information Retrieval, ISMIR’03, 2003.
[14] M. Grachten, J. L. Arcos and R. L. Mántaras, “Melodic Similarity: Looking for a Good Abstraction Level,” Proc. of International Symposium on Music Information Retrieval,
ISMIR’04, 2004.
[15] K. Hirata and S. Matsuda, “Interactive Music Summarization based on GTTM,” Proc. of
International Symposium on Music Information Retrieval, ISMIR’02, 2002.
[16] J. L. Hsu, “Content-based Music Information Retrieval and Analysis,” Ph.D. Thesis, Department of Computer Science, National Tsing Hua University, Taiwan, 2001.
[17] J. L. Hsu, C. C. Liu, and A. L. P. Chen, “Discovering Nontrivial Repeating Patterns in Music Data,” IEEE Transactions on Multimedia, Vol. 3, No. 3, pp. 311-325, 2001.
[18] J. L. Hsu, C. C. Liu, and A. L. P. Chen, “Efficient Repeating Pattern Finding in Music Database,” Proc. of Conference on Information and Knowledge Management, CIKM’98, 1998.
[19] C. S. Iliopoulos, K. Lemström, M. Nuyad, and Y. J. Pinzón, “Evolution of Musical Motifs in Polyphonic Passages,” Proc. of Symposium on AI and Creativity in Arts and Science,
AISB’02, 2002.
[20] C. S. Iliopoulos, T. Lecroq, L. Mouchard, and Y. Z. Pinzon, “Computing Approximate Repetitions in Musical Sequence,” International Journal of Computer Mathematics, Vol.3,
No. 1&2, 2002.
[21] J. S. R. Jang, H. R. Lee, C. H. Yeh, “A Query-by-Tapping System for Music Retrieval,”
Proc. of IEEE Pacific-Rim Conference on Multimedia, 2001.
[22] M. H. Jian, C. H. Lin, and A. L. P. Chen, “Perceptual Analysis for Music Segmentation,”
Proc. SPIE Storage and Retrieval Methods and Applications for Multimedia, 2004.
[23] I. Koprinska and S. Carrato, “Temporal Video Segmentation: A Survey,” Signal Processing
Image Communication, Vol. 16, No. 5, pp. 477-500, 2001.
[24] F. F. Kuo, “Melody Style Mining Using Chord Features,” Master Thesis, Department of Computer Science, National Chengchi University, Taiwan, 2003.
[25] O. Lartillot, “Discovering Musical Patterns through Perceptive Heuristics,” Proc. of
International Symposium on Music Information Retrieval, ISMIR’03, 2003.
[26] O. Lartillot, S. Dubnov, G. Assayag, and G. Bejerano, “Automatic Modeling of Music Style,” Proc. Of International Computer Music Conference, ICMC, 2002.
[27] F. Lerdahl and R. Jackendoff, A Generative Theory of Tonal Music, MIT Press, 1983.
[28] C. C. Liu, J. L. Hsu, and A. L. P. Chen, “Efficient Theme and Non-Trivial Repeating Pattern Discovering in Music Databases,” Proc. of International Conference on Data
Engineering, ICDE’99, 1999.
[29] B. Logan and S. Chu, “Music Summarization Using Key Phrases,” Proc. of International
Conference on Speech, Acoustics, and Signal Processing, ICASSP’00, 2000.
[30] L. Lu, H. You, and H.J. Zhang, “A New Approach to Query by Humming In Music Retrieval,” Proc. of IEEE International Conference on Multimedia and Expo, ICME’01, 2001.
[31] K. D. Martin, E. D. Scheirer, and B. L. Vercoe, “Music content analysis through Models of Audition,” Proc. of ACM Multimedia Workshop on Content-Based Processing of Music, 1998.
[32] C. Meek and W. P. Birmingham, “Automatic Thematic Extractor,” Proc. of International
Symposium on Music Information Retrieval, ISMIR’01, 2001.
[33] C. Meek and W. P. Birmingham, “Automatic Thematic Extractor,” Journal of Intelligent
Information Systems, Vol. 21, No. 1, pp. 9-33, 2003.
[34] D. Meredith, K. Lemström, and G. A. Wiggins, “Algorithm for Discover Repeating Patterns in Multidimensional Representations of Polyphonic Music,” Journal of New Music
Research, Vol.31, No.4, pp. 321-345,
2002.[35] J. Pickens, “Feature Selection for Polyphonic Music Retrieval,” Proc. of the ACM
International Conference on Research and Development in Information Retrieval, SIGIR’01,
2001.[36] D. Rizo, J. M. Iñesta, F. Moreno-Seco, “Tree-structured Representation of Musical Information,” Iberian Conference on Pattern Recognition and Image Analysis, 2003.
[37] M. K. Shan, F. F. Kuo, and M. F. Chen, “Music Style Mining and Classification by Melody,” Proc. of IEEE International Conference on Multimedia and Expo, ICME’02, 2002.
[38] M. K. Shan and F. F. Kuo, “Music Style Mining and Classification by Melody,” IEICE
Transactions on Information and System, Vol. E86-D, No. 4, 2003.
[39] H. H. Shih, S. S. Narayanan, and C. C. J. Kuo, “A Dictionary Approach to Repeating Pattern Finding in Music,” Proc. of IEEE International Conference on Multimedia and
Expo, ICME’01, 2001.
[40] C. Spevak, B. Thom, and Karin Höthker, “Evaluating Melodic Segmentation,” Proc. of
International Conference on Music and Artificial Intelligence, ICMAI’02, 2002.
[41] L. Stein, Structure & Style：The Study and Analysis of Musical Forms, Summy-Birchard Music, 1979.
[42] D. Temperley, The Cognition of Basic Musical Structires, MIT press, 2001.
[43] B. Thom, “BoB: An Improvisational Music Companion,” Ph.D. Thesis, Department of Computer Science, Carnegie Mellon University, USA, 2001.
[44] B. Thom, C. Spevak, and K. Höthker, “Melodic Segmentation: Evaluating the Performance of Algorithms and Musical Experts,” Proc. of the International Computer Music Conference,
ICMC’02, 2002.
[45] Y. H. Tseng, “Content-Based Retrieval for Music Collections,” Proc. of ACM Conference
on Research and Development in Information Retrieval, SIGIR’99, 1999.
[46] A. L. Uitdenbogerd and J. Zobel, “Manipulation of Music For Melody Matching,” Proc. of
ACM International Conference on Multimedia, MM’98, 1998.
[47] A. L. Uitdenbogerd and J. Zobel, “Melodic Matching Techniques for Large Music Databases,” Proc. of ACM International Conference on Multimedia, MM’99, 1999.
[48] T. Weyde, “Integrating Segmentation and Similarity in Melodic Analysis,” Proc. of the
International Conference on Music Perception and Cognition, ICMPC’02, 2002.
[49] C. Xu, Y. Zhu, and Q. Tian, “Automatic Music Summarization Based on Temporal, Spectral and Cepstral Features,” Proc. of IEEE International Conference on Multimedia and Expo,
ICME’02, 2002.
[50] M. Zinn and R. Hogenson, Basic of Music: Opus 1, Schirmer books, 1994.
計畫成果自評
就研究內容與原計畫相符程度、達成預期目標情況
本研究計畫已經完成音樂分段技術，以提供音樂瀏覽之技術研究。本計畫利用音樂的 動機輔助對音樂作主題式的分段，進一步提供使用者以主題段落瀏覽音樂。由於音樂的主 題必須透過音樂內容做分析，所以我們探勘出音樂的動機，改善音樂主題分段的效果。此 外，我們提出一個新的評估分段效果的方法，改進以前只考慮Precision 或 Recall 所可能造 成的缺點。本計畫因為實驗所需之音樂，在音樂轉譯時品質不佳，因此申請計畫延期。在 延期三個月中，我們已經完成音樂的實驗。實驗結果顯示我們所研究的音樂分段效果準確 率最高可以到84%，平均 65%。
研究成果之學術或應用價值
本計畫的研究成果，在學術價值方面，我們提出新的音樂動機探勘的演算法。我們也提出 split and merge 的方法來找出音樂的主題片段。本研究成果也將有助於音樂結構分析(music structure analysis)與音樂摘要產生(music summarization)的相關研究。 的IEEE Syetems, Man and Cybernetics (IEEE SMC)。此外，我們也預計將部分研究成果整理 後，投稿至2007 年的 ACM/IEEE Joint Conference on Digital Libraries (JCDL)。我們也預計 將這些研究成果整理投稿至學術期刊。
可供推廣之研發成果資料表
□ 可申請專利 □ 可技術移轉 日期：95 年 5 月 31 日
國科會補助計畫
計畫名稱：數位音樂典藏之資料探勘與智慧型檢索技術 計畫主持人：沈錳坤
計畫編號：NSC 94-2422-H-004-003- 學門領域：資訊工程
技術/創作名稱
以音樂動機為基礎之音樂分段技術We proposed the theme segmentation for digital music.
There are four steps in the theme segmentation technique. Firstly, we extract main melody from original music. In the second step, rough segments are generated from main melody by mining non-trivial repeating patterns. Then, motives are detected from rough segments.
We modify the mining algorithm for discovering frequent patterns by applying motivic treatment rules proposed by Stein. Finally, we segment main melody based on the generated motives.