A research on automatic music performance rendering system and automatic accompaniment system

2016 Fiscal Year Final Research Report

• Project/Area Number: 26730182
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Entertainment and game informatics 1
• Research Institution: Nagoya Institute of Technology
• Principal Investigator: SAKO SHINJI 名古屋工業大学, 工学(系)研究科(研究院), 助教 (30396791)
• Project Period (FY): 2014-04-01 – 2017-03-31
• Keywords: 演奏表情生成 / 演奏追跡 / 楽譜アライメント / 条件付き確率場 / 演奏分析 / 自動伴奏再生

Outline of Final Research Achievements

In this research, we propose a framework for expressive performance rendering system that learns characteristics in musical instrument performance from real performance data, reproduces (transfers) their performance style. In other side, we also proposed advanced technique for automatic alignment between acoustic signal and music score (score following). This technique also would be useful in the research on expressive performance rendering by generating the dataset. Regarding the expressive performance rendering system we demonstrated that it is possible to analyze and visualize the characteristic performances as well as work on improving the performance generation model such as improvement of reproducibility. Regarding score alignment, we mainly work on improving the accuracy of score alignment. Our results shows an effectiveness of the use of higher order musical information in music score in addition to the basic information such as note sequence.

Free Research Field

音楽情報科学