Our recent paper, “Green MIR?: Investigating computational cost of recent music-Ai research in ISMIR” was nominated for the best paper award at ISMIR 2024 in San Francisco. In this work, we study the environmental footprint of automatic music generation (AMG) and computing-heavy music analysis at the ISMIR conferences 2017–2023. Our results demonstrate a widespread lack of transparency in model training documentation and provide the first estimates of energy consumption related to model training at ISMIR.

Read more at https://zenodo.org/records/14877351.