Presentation 2024-03-13
Drum transcription based on periodicity between bars
Masaki Suga, Tetsuya Matsumoto, Yoshinori Takeuchi, Hiroaki Kudo,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Automatic musical score estimation from acoustic signals has been studied for a long time. In this study, acoustic signals are used as input and neural networks are used to estimate the onset probability of instruments of drums at each time. The learning process is regularized based on the autocorrelation function, and the feature of the learning process is that the same sequences appear in each bar. For the assumption that the same sequences appear in every two bars, we used a frame-based method based on the conventional method, and a regularization method based on the periodicity of the acoustic signal. Tatum variation and regularization improved the F-measure by about 0.1 compared to the frame-based method. The regularization based on the repetition structure of music showed improvement in F-measure when the repetitions were in odd number bars.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Drum transcription / Acoustic signal processing / CNN / periodicity
Paper # IMQ2023-27,IE2023-82,MVE2023-56
Date of Issue 2024-03-06 (IMQ, IE, MVE)

Conference Information
Committee IE / MVE / CQ / IMQ
Conference Date 2024/3/13(3days)
Place (in Japanese) (See Japanese page)
Place (in English) Okinawa Sangyo Shien Center
Topics (in Japanese) (See Japanese page)
Topics (in English) Media of five senses, Multimedia, Media experience, Picture codinge, Image media quality, Network,quality and reliability, etc(AC)
Chair Hiroyuki Bandoh(NTT) / Kiyoshi Kiyokawa(NAIST) / Takefumi Hiraguri(Nippon Inst. of Tech.) / Hiroaki Kudo(Nagoya Univ.)
Vice Chair Yuichi Tanaka(Osaka Univ.) / Toshihiko Yamazaki(Univ. of Tokyo) / Sumaru Niida(KDDI Research) / Takahiro Matsuda(Tokyo Metropolitan Univ.) / Gou Hasegawa(Tohoku Univ.) / Sumaru Niida(KDDI Research) / Gosuke Ohashi(Shizuka Univ.)
Secretary Yuichi Tanaka(NHK) / Toshihiko Yamazaki(Tottori Univ.) / Sumaru Niida(Otsuma Women's Univ.) / Takahiro Matsuda(DNP) / Gou Hasegawa(NTT) / Sumaru Niida(NTT) / Gosuke Ohashi(Tama Univ.)
Assistant Kazunori Uruma(Kogakuin Univ.) / Shinobu Kudo(KDDI Research) / Hidehiko Shishido(Univ. of Tsukuba) / Atsushi Nakazawa(Kyoto Univ.) / Naoya Tojo(KDDI Research) / Naoki Hagiyama(NTT) / Yuji Tatada(Univ. of Tokyo) / Ryo Nakamura(Fukuoka Univ.) / Toshiro Nakahira(NTT) / Kenta Tsukatsune(Okayama Univ. of Science) / Kuniharu Imai(Nagoya Univ.) / Takashi Yamazoe(Seikei Univ.)

Paper Information
Registration To Technical Committee on Image Engineering / Technical Committee on Media Experience and Virtual Environment / Technical Committee on Communication Quality / Technical Committee on Image Media Quality
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Drum transcription based on periodicity between bars
Sub Title (in English)
Keyword(1) Drum transcription
Keyword(2) Acoustic signal processing
Keyword(3) CNN
Keyword(4) periodicity
1st Author's Name Masaki Suga
1st Author's Affiliation Nagoya University(Nagoyoa Univ.)
2nd Author's Name Tetsuya Matsumoto
2nd Author's Affiliation Nagoya University(Nagoyoa Univ.)
3rd Author's Name Yoshinori Takeuchi
3rd Author's Affiliation Daido University(Daido Univ.)
4th Author's Name Hiroaki Kudo
4th Author's Affiliation Nagoya University(Nagoyoa Univ.)
Date 2024-03-13
Paper # IMQ2023-27,IE2023-82,MVE2023-56
Volume (vol) vol.123
Number (no) IMQ-430,IE-432,MVE-433
Page pp.pp.81-86(IMQ), pp.81-86(IE), pp.81-86(MVE),
#Pages 6
Date of Issue 2024-03-06 (IMQ, IE, MVE)