Associate Professor, ICE Lab.
Graduate School of Information Science and Electrical Engineering, Kyushu University
nakamura[at]inf.kyushu-u.ac[dot]jp
Publications and Talks [arXiv] [Google Scholar]
Curriculum Vitae
Graduate School of Information Science and Electrical Engineering, Kyushu University
nakamura[at]inf.kyushu-u.ac[dot]jp
Publications and Talks [arXiv] [Google Scholar]
Curriculum Vitae
News
-
I have moved to Kyushu University.
-
Award
We received a Best Paper Award at CMMR 2023.
-
Award
We received a Best Paper Award (2nd place) at APSIPA ASC 2023.
-
Paper
Our paper “Dynamic cluster structure and predictive modelling of music creation style distributions” is published in Royal Society Open Science.
-
Dataset
Dataset of melody statistics of Japanese enka music is published.
-
Release
“If you can't score music, just beat it.” An article on piano transcription is released on the Kyoto University webpage.
-
Web system
CREEVO, an evolutionary system of automatic composition, is open now. Melodies of a variety of styles can be generated for Japanese lyrics users input. This is a piece of ongoing work.
-
Paper
Our paper “Non-local musical statistics as guides for audio-to-score piano transcription” is published in Information Sciences.
-
Dataset
PIG (piano fingering) dataset is published. The dataset is freely available for research purposes.
Research Interests
- Mathematical models of intelligence
- Cultural evolution
- Statistical learning and machine learning
- Statistical physics and evolutionary dynamics
- Music information processing
Research Topics
Evolution of music
Music styles change over time. How and why do they change? Can we predict future music styles? We are studying the evolution of music based on computational analyses of music data and theoretical models of music creation process and social selection.
Automatic music generation
Can we compose music by computers? We are studying music creation from the viewpoints of statistical learning and optimization. We are interested in how individual music styles are developed and transmitted, and in the roles of active and dynamic statistical learning in musical creativity.Melody style conversion Piano arrangement
Music transcription
Trained musicians can listen to music audio and represent it as a musical score. We are studying computational methods to emulate this ability. Automatic music transcription is a key experimental method for analyzing music audio in the symbolic domain.
Evaluation metrics (MUSTER)
Automatic music accompaniment
Can computers listen to music and interact with humans? We are studying a method for understanding human performances and playing accompaniments with humans. Various aspects of human performances including tempo changes, errors, repeats and skips, ornaments are studied.Eurydice system
Piano fingering
Fingering is a basic skill for pianists, but the principle of finding appropriate fingerings is not fully understood scientifically. We are constructing computational models to formulate and search for optimal fingerings and to measure performance difficulty.
Symbolic music alignment
Study for comparing two symbolic representations of a musical piece (for example a MIDI recording and a corresponding musical score) at the note level.Software and demo
Selected Publications and Talks
(See the complete list)
- Tengyu Deng, Eita Nakamura, Ryo Nishikimi, Kazuyoshi Yoshii
End-to-End Singing Transcription Based on CTC and HSMM Decoding with a Refined Score Representation
APSIPA Transactions on Signal and Information Processing, to appear, 2024.
- Eita Nakamura, Yasuyuki Saito
Estimation of creator influences based on cultural evolution models of color styles in painting arts (in Japanese)
The Journal of the Institute of Image Electronics Engineers of Japan, Vol. 53, 1, pp. 19-27, Feburary 2024.
- Eita Nakamura
Recent developments and open problems in audio-to-score music transcription
Talk at the Eighth International Workshop on Symbolic-Neural Learning (SNL2024), Tokyo, Japan, 26/June/2024. - Ryota Nakajima, Arata Shirakami, Hayato Tsumura, Kouki Matsuda, Eita Nakamura, Masanori Shimono
Mutual generation in neuronal activity across the brain via deep neural approach, and its network interpretation
Communications Biology, Vol. 6, 1105, 2023. - Takuto Nabeoka, Eita Nakamura, Kazuyoshi Yoshii
Automatic orchestration of piano scores for wind bands with user-specified instrumentation
Proc. 16th International Symposium on Computer Music Multidisciplinary Research (CMMR), pp. 387-394, November 2023. - Eita Nakamura, Tim Eipert, Fabian C. Moss
Historical changes of modes and their substructure modeled as pitch distributions in plainchant from the 1100s to the 1500s
Proc. 16th International Symposium on Computer Music Multidisciplinary Research (CMMR), pp. 450-461, November 2023. - Eita Nakamura
Computational analysis of selection and mutation probabilities in the evolution of chord progressions
Proc. 16th International Symposium on Computer Music Multidisciplinary Research (CMMR), pp. 462-473, November 2023. (Best Paper Award) - Daichi Kamakura, Takehisa Ooyama, Eita Nakamura, Kazuyoshi Yoshii
Joint drum transcription and metrical analysis based on periodicity-aware multi-task learning
Proc. 15th Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA), pp. 145-151, November 2023. - Daichi Kamakura, Eita Nakamura, Kazuyoshi Yoshii
CTC2: End-to-end drum transcription based on connectionist temporal classification with constant tempo constraint
Proc. 15th Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA), pp. 152–158, November 2023. - Eita Nakamura, Yasuyuki Saito
Evolutionary analysis and cultural transmission models of color style distributions in painting arts
Proc. 15th Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA), pp. 493–500, November 2023. - Tengyu Deng, Eita Nakamura, Kazuyoshi Yoshii
Audio-to-score singing transcription based on joint estimation of pitches, onsets, and metrical positions with tatum-level CTC loss
Proc. 15th Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA), pp. 570–577, November 2023. (Best Paper Award (2nd place)) - Norihiro Kato, Eita Nakamura, Kyoko Mine, Orie Doeda, Masanao Yamada
Computational analysis of audio recordings of piano performance for automatic evaluation
Proc. 18th European Conference on Technology Enhanced Learning (ECTEL), pp. 586–592, September 2023. - Eita Nakamura, Hitomi Kaneko, Takayuki Itoh, Kunihiko Kaneko
Experimental evolution of music styles using automatic composition models
Proc. 2023 Conference on Artificial Life (ALIFE), pp. 660–662, July 2023. - Moyu Terao, Eita Nakamura, Kazuyoshi Yoshii
Neural band-to-piano score arrangement with stepless difficulty control
Proc. 48th IEEE International Conference on Acoustics, Speech, and Signal Processing Conference (ICASSP), 1415, pp. 1-5, June 2023.2022 and before
- Rajsuryan Singh, Eita Nakamura
Dynamic cluster structure and predictive modelling of music creation style distributions 
Royal Society Open Science, Vol. 9, 220516, 2022. [arXiv:2205.13923] - Eita Nakamura
Conjugate Distribution Laws in Cultural Evolution via Statistical Learning
Physical Review E, Vol. 104, 034309, 2021. [arXiv:2102.01465] - Kentaro Shibata, Eita Nakamura, Kazuyoshi Yoshii
Non-Local Musical Statistics as Guides for Audio-to-Score Piano Transcription
Information Sciences, Vol. 566, pp. 262-280, 2021. [arXiv:2008.12710] - Eita Nakamura, Yasuyuki Saito, Kazuyoshi Yoshii
Statistical Learning and Estimation of Piano Fingering
Information Sciences, Vol. 517, pp. 68-85, 2020. [arXiv:1904.10237] - Eita Nakamura, Kunihiko Kaneko
Statistical Evolutionary Laws in Music Styles
Scientific Reports, Vol. 9, No. 15993, pp. 1-11, 2019. [arXiv:1809.05832] - Eita Nakamura, Kazuyoshi Yoshii
Statistical Piano Reduction Controlling Performance Difficulty
APSIPA Transactions on Signal and Information Processing, Vol. 7, No. e13, pp. 1–12, 2018. [arXiv:1808.05006]
- Eita Nakamura, Kazuyoshi Yoshii, Haruhiro Katayose
Performance Error Detection and Post-Processing for Fast and Accurate Symbolic Music Alignment
Proc. 18th International Society for Music Information Retrieval Conference (ISMIR), pp. 347-353, 2017. - Eita Nakamura, Kazuyoshi Yoshii, Shigeki Sagayama
Rhythm Transcription of Polyphonic Piano Music Based on Merged-Output HMM for Multiple Voices
IEEE/ACM Transactions on Audio, Speech and Language Processing, Vol. 25, No. 4, pp. 794-806, 2017. [arXiv:1701.08343]
- Eita Nakamura, Nobutaka Ono, Shigeki Sagayama, Kenji Watanabe
A Stochastic Temporal Model of Polyphonic MIDI Performance with Ornaments
Journal of New Music Research, Vol. 44, No. 4, pp. 287-304, 2015. [arXiv:1404.2314]
- Koichi Hamaguchi, Eita Nakamura, Satoshi Shirai, Tsutomu T. Yanagida
Decaying Dark Matter Baryons in a Composite Messenger Model
Physics Letters B, Vol. 674, pp. 299-302, 2009. [arXiv:0811.0737]
2023 and later
(See the complete list)
Contact
Eita NakamuraWest 4 Room 305, Motooka 744, Nishi-ku, Fukuoka 819-0395, Japan
e-mail: nakamura[at]inf.kyushu-u.ac[dot]jp
phone: +81-92-802-3808