In this paper, we propose a method that estimates appropriate violin fingering according to the performer’s skill level based on a conditional random field (CRF). A violin is an instrument that can produce the same pitch for different fingering patterns, and these patterns depend on skill level. We previously proposed a statistical method for violin fingering estimation, but that method required a certain amount of training data in the form of fingering annotation corresponding to each note in the music score. This was a major issue of our previous method, because it takes time and effort to produce the annotations. To solve this problem, we proposed a method to automatically generate training data for a fingering model using existing violin textbooks. Our experimental results confirmed the effectiveness of the proposed method.
CITATION STYLE
Sako, S., Nagata, W., & Kitamura, T. (2015). Violin fingering estimation according to the performer’s skill level based on conditional random field. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9170, pp. 485–494). Springer Verlag. https://doi.org/10.1007/978-3-319-20916-6_45
Mendeley helps you to discover research relevant for your work.