イトウ　アキノリ   Akinori Ito   伊藤　彰教    所属   メディア学部 メディア学科    職種   特任講師
言語種別	英語
発行・発表の年月	2023/11
形態種別	国際会議論文
査読	査読あり
標題	A Research on Music Generation by Deep-Learning including ornaments - A case study of world harp instruments -
執筆形態	共著
掲載誌名	Proceedings of CMMR2023
掲載区分	国外
出版社・発行元	CMMR 2023 The 16th International Symposium on Computer Music Multidisciplinary Resarch
巻・号・頁	pp.732-735
総ページ数	4
著者・共著者	Arturo Alejandro Arzamendia Lopez, Akinori Ito, Koji Mikami
概要	In this research, we explore the application of deep learning techniques, including recurrent neural networks and LSTM, to traditional Paraguayan music known as "Guarania". This style is characterized by specific playing techniques and ornaments such as arpeggios and glissandos, which are executed using the Paraguayan harp. The learning and generation processes are performed individually using the TensorFlow and Keras libraries comparing the different results from different architectures to identify which one generates the most accurate or similar harp music that captures the intricacies of "Guarania" style. Furthermore, in future work, we demonstrate the capability of these techniques in other world harp music styles, such as Meiji period music employing the Koto, and western music from the 19th and 20th centuries incorporating the concert harp.
外部リンクURL	https://zenodo.org/records/10115215