We propose a visual approach for interactive, AI-assisted composition that serves as a compromise between fully automatic and fully manual composition. Instead of generating a whole piece, the AI takes on the role of an assistant that generates short melodies for the composer to choose from and adapt. In an iterative process, the composer queries the AI for continuations or alternative fill-ins, chooses a suggestion, and adds it to the piece. As listening to many suggestions would take time, we explore different ways to visualize them, to allow the composer to focus on the most interesting-looking melodies. We also present the results of a qualitative evaluation with five composers.
%0 Conference Paper
%1 simeon_rau_2022_7316618
%A Rau, Simeon
%A Heyen, Frank
%A Wagner, Stefan
%A Sedlmair, Michael
%B Proceedings of the 23rd International Society for Music Information Retrieval Conference
%C Bengaluru, India
%D 2022
%I ISMIR
%K myown from:simeonrau iste-se:wagnerst visus:heyenfk cybervalley peerreviewed visus visus:rausn vis(us) visus:sedlmaml iste-se
%P 151-159
%R 10.5281/zenodo.7316618
%T Visualization for AI-Assisted Composing
%U https://doi.org/10.5281/zenodo.7316618
%X We propose a visual approach for interactive, AI-assisted composition that serves as a compromise between fully automatic and fully manual composition. Instead of generating a whole piece, the AI takes on the role of an assistant that generates short melodies for the composer to choose from and adapt. In an iterative process, the composer queries the AI for continuations or alternative fill-ins, chooses a suggestion, and adds it to the piece. As listening to many suggestions would take time, we explore different ways to visualize them, to allow the composer to focus on the most interesting-looking melodies. We also present the results of a qualitative evaluation with five composers.
@inproceedings{simeon_rau_2022_7316618,
abstract = {We propose a visual approach for interactive, AI-assisted composition that serves as a compromise between fully automatic and fully manual composition. Instead of generating a whole piece, the AI takes on the role of an assistant that generates short melodies for the composer to choose from and adapt. In an iterative process, the composer queries the AI for continuations or alternative fill-ins, chooses a suggestion, and adds it to the piece. As listening to many suggestions would take time, we explore different ways to visualize them, to allow the composer to focus on the most interesting-looking melodies. We also present the results of a qualitative evaluation with five composers.},
added-at = {2023-01-26T17:58:45.000+0100},
address = {Bengaluru, India},
author = {Rau, Simeon and Heyen, Frank and Wagner, Stefan and Sedlmair, Michael},
biburl = {https://puma.ub.uni-stuttgart.de/bibtex/297146b51e08cfa3c0359c203dcc0d0f7/visus},
booktitle = {Proceedings of the 23rd International Society for Music Information Retrieval Conference},
doi = {10.5281/zenodo.7316618},
interhash = {c24850e587da15581e74fbff32b1f6bb},
intrahash = {97146b51e08cfa3c0359c203dcc0d0f7},
keywords = {myown from:simeonrau iste-se:wagnerst visus:heyenfk cybervalley peerreviewed visus visus:rausn vis(us) visus:sedlmaml iste-se},
month = dec,
pages = {151-159},
publisher = {ISMIR},
timestamp = {2023-01-27T10:35:30.000+0100},
title = {Visualization for AI-Assisted Composing},
url = {https://doi.org/10.5281/zenodo.7316618},
venue = {Bengaluru, India},
year = 2022
}
