AI DJ Project — A dialogue between AI and a human
DJ Performance Based Upon A Dialogue between Performer and AI
"The AI DJ Project - A Dialogue Between AI and a Human" is a project that explores the unique dialogue between humans and AI through a "Back to Back" playing style, where a human DJ and an AI DJ take turns selecting music alternately.
A DJ is a person who mixes different sources of pre-existing recorded music, usually for the audience. It is regarded as a highly creative process to select appropriate music and mix them in smooth and pleasant ways.
AI DJ uses three neural networks that predict music genres, instruments, and drum machines to select the next song and match the beat using turntables. While analyzing the selection of human DJs, it quantifies the audience’s reaction with motion tracking technology and uses it to select the next song. AI DJ sometimes selects music that matches the live audience’s energy, and other times it makes unexpected selections that lead to unpredictable actions, bringing an interesting tension to the performance and inspiring human DJ.
The aim of this project is not to automate the entire DJ process. Rather, it aims to explore new forms of expression that arise from the collaboration between humans and AI, by highlighting the various possibilities achievable through technology and human creativity.
Video: Yasuhiro Tani, Courtesy of Yamaguchi Center for Arts and Media [YCAM] / Edit: Qosmo, inc.
Performances
2016/9/4 | 2045 Generation #4（京都岡崎音楽祭「OKAZAKI LOOPS」内）| 京都国立近代美術館 2016/10/27 | 2045 × LIFE PAINT Supported by VOLVO CAR JAPAN | 代官山UNIT 2017/2/17 | DIGITAL CHOC — マシン・デジラント 欲望する機械 — | 渋谷WWW 2017/9/14 | Festival Speculum Artium 2017 in Slovenia | Zavod za kulturo Delavski dom Trbovlje, Slovenia 2017/9/21 | SCOPITONE FESTIVAL 2017 in France | STEREOLUX, Nantes, France 2017/11/3 | MUTEK.JP | 日本科学未来館（Miraikan） 2017/12/15 | sound tectonics #20（Guest DJ : tofubeats, Licaxxx） | 山口情報芸術センター[YCAM] 2019/5/7 | Google I/O 2019 | Mountain View, California, USA 2019/6/1 | Japan Media Arts Festival x MUTEK.JP | 日本科学未来館（Miraikan）
References
[1] AlgoRhythm — Neukom Institute Turing Tests in the Creative Arts. - AlgoRhythm [2] Keunwoo Choi, Gyrgy Fazekas and Mark Sandler. Transfer learning for music classification and regression tasks. Mar 2017. [3] Matthew E. P. Davies and Sebastian Böck. Evaluating the Evaluation Measures for Beat Tracking. ISMIR, 2014. [4] Daito Manabe and Nao Tokui. 2045 AI DJ Party. - 2045 AI DJ Party
Articles
- AI DJ Project — A dialog between human and AI through music
Credits
Concept / Programming: Nao Tokui (Qosmo, Inc.) Visualization: Shoya Dozono (Qosmo, Inc.) Database: Robin Jungers (Qosmo, Inc.) Assistant: Yuma Kajihara (Qosmo, Inc.), Ryosuke Nakajima (Qosmo, Inc.), Makoto Amano (Qosmo, Inc.) Robot: TASKO inc. Project Management: Miyu Hosoi (Qosmo, Inc.) Customized turntable for AI: Mitsuhito Ando (YCAM) Project Support: YCAM InterLab