σGTTM System

σGTTM combines the generative theory of tonal music (GTTM) and statistical learning. We previously devised exGTTM, which has accommodated the original GTTM to computer implementations. The exGTTM has adjustable parameters, these parameters have to be manually configured. Therefore, it is not perfectly suited for automation. To make complete automation possible, we combined statistical learning with GTTM to create σGTTM. To prevent the sparseness problem, we abstracted data properly by using the GTTM rules for analyzing musical structures. We use the abstracted data to construct a decision tree, which is a model of decisions and their possible consequences. With σGTTM, we can segment a melody automatically from the decision tree, dependent on conditional probability. Experimental results showed that σGTTM outperformed the baseline exGTTM.

pa dt2

References

  1. Kouhei Kanamori, Junichi Hoshino, Masatoshi Hamanaka: “System to Analyze Music Structure Based on Music Theory GTTM Using Clustering and Machine Learning”, IPSJ Special Interest Group on Music and Computer 2016-MUS-110-18, Vol. 2016, No.18, March 2016. [PDF]
  2. Masatoshi Hamanaka, Keiji Hirata, Satoshi Tojo: “Sigma GTTM III: Learning based Time-span Tree Generator based on PCFG”, Peer-reviewed, Proceedings of The 11th International Symposium on Computer Music Multidisciplinary Research (CMMR 2015), pp.303-317, June 16-19, 2015. [PDF]
  3. Masatoshi Hamanaka: “Developing σGTTM III”, JSAI2015 The 29th Annual Conference of the Japanese Society for Artificial Intelligence, 2C5-OS-21b-1, May 2015. [PDF]
  4. Kouhei Kanamori, Masatoshi Hamanaka, Junichi Hoshino: “Method to detect GTTM Local Grouping Boundaries based on Clustering and Statistical Learning”, Proceedings of the 42nd International Computer Music Conference (ICMC) joint with the 13rd Sound & Music Computing conference (SMC), October 2014. [PDF]
  5. Kouhei Kanamori, Masatoshi Hamanaka: “Music Theory based on Clustering and Statistical Learning σGTTMⅡ: Detection of Local Grouping Boundary”, IPSJ Special Interest Group on Music and Computer, 2014-MUS-102, No.2, 7 pages, February 2014. [PDF]
  6. Yuji Miura, Masatoshi Hamanaka, Keiji Hirata, Satoshi Tojo: Use of Decision Tree to Detect GTTM Group Boundaries, Proceedings of the 2009 International Computer Music Conference (ICMC2009), pp. 125-128, August 2009. [PDF]
  7. 三浦右士, 浜中雅俊, 平田圭二, 東条敏: “統計的学習に基づく音楽理論σGTTM：局所的グルーピング境界の検出”, 情報処理学会 音楽情報科学研究会 研究報告 2008-MUS-76-14, Vol.2008, No.78, pp. 75-82, August 2008. [PDF]