Presentation 2022-03-03
Few-Shot Music Artist Classification
Tianshuai Yu, Yoshimasa Tsuruoka,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Music artist classification is known as a task in the field of Music Information Retrieval(MIR). Recently, due to the improvement of calculation efficiency, it becomes possible to solve the music artist classification task using deep learning. However, in real-world cases, labeled music pieces of each artist might be limited, making the task a few-shot learning task. Unfortunately, hardly any research about few-shot music artist classification has been reported. In this research, we combine a music artist classification method using CRNN and several few-shot learning algorithms well known in the field of image classification to tackle with the few-shot music artist classification task. In our experiment, Matching Network, Prototypical Network and Relation Network are shown effective. However, Baseline++, which is reported to be as effective as mentioned few-shot learning methods, performs poorly in the few-shot music artist classification task. This shows the differences between the domain of music artist classification and the domain of image classification.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Music Information Retrieval / Music Artist Classification / Few-Shot Learning
Paper # SIS2021-36
Date of Issue 2022-02-24 (SIS)

Conference Information
Committee SIS
Conference Date 2022/3/3(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Online
Topics (in Japanese) (See Japanese page)
Topics (in English) Soft Computing, etc.
Chair Noriaki Suetake(Yamaguchi Univ.)
Vice Chair Tomoaki Kimura(Kanagawa Inst. of Tech.) / Naoto Sasaoka(Tottori Univ.)
Secretary Tomoaki Kimura(NTT) / Naoto Sasaoka(National Inst. of Tech., Ube College)
Assistant Soh Yoshida(Kansai Univ.) / Yoshiaki Makabe(Kanagawa Inst. of Tech.)

Paper Information
Registration To Technical Committee on Smart Info-Media Systems
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Few-Shot Music Artist Classification
Sub Title (in English)
Keyword(1) Music Information Retrieval
Keyword(2) Music Artist Classification
Keyword(3) Few-Shot Learning
1st Author's Name Tianshuai Yu
1st Author's Affiliation Tokyo University(Tokyo Univ.)
2nd Author's Name Yoshimasa Tsuruoka
2nd Author's Affiliation Tokyo University(Tokyo Univ.)
Date 2022-03-03
Paper # SIS2021-36
Volume (vol) vol.121
Number (no) SIS-398
Page pp.pp.32-37(SIS),
#Pages 6
Date of Issue 2022-02-24 (SIS)