{"ID":2874319,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.05256","arxiv_id":"2509.05256","title":"Recomposer: Event-roll-guided generative audio editing","abstract":"Editing complex real-world sound scenes is difficult because individual sound sources overlap in time. Generative models can fill-in missing or corrupted details based on their strong prior understanding of the data domain. We present a system for editing individual sound events within complex scenes able to delete, insert, and enhance individual sound events based on textual edit descriptions (e.g., ``enhance Door'') and a graphical representation of the event timing derived from an ``event roll'' transcription. We present an encoder-decoder transformer working on SoundStream representations, trained on synthetic (input, desired output) audio example pairs formed by adding isolated sound events to dense, real-world backgrounds. Evaluation reveals the importance of each part of the edit descriptions -- action, class, timing. Our work demonstrates ``recomposition'' is an important and practical application.","short_abstract":"Editing complex real-world sound scenes is difficult because individual sound sources overlap in time. Generative models can fill-in missing or corrupted details based on their strong prior understanding of the data domain. We present a system for editing individual sound events within complex scenes able to delete, in...","url_abs":"https://arxiv.org/abs/2509.05256","url_pdf":"https://arxiv.org/pdf/2509.05256v1","authors":"[\"Daniel P. W. Ellis\",\"Eduardo Fonseca\",\"Ron J. Weiss\",\"Kevin Wilson\",\"Scott Wisdom\",\"Hakan Erdogan\",\"John R. Hershey\",\"Aren Jansen\",\"R. Channing Moore\",\"Manoj Plakal\"]","published":"2025-09-05T17:14:29Z","proceeding":"cs.SD","tasks":"[\"cs.SD\",\"cs.AI\",\"cs.LG\",\"eess.AS\"]","methods":"[\"Transformer\"]","has_code":false}