An encoder-decoder transformer conditions on text actions and a time-aligned event roll to delete, insert, or enhance individual sound events in dense audio scenes.
Text-to-audio generation using instruction guided latent diffusion model,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.SD 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
support 1representative citing papers
citing papers explorer
-
Recomposer: Event-roll-guided generative audio editing
An encoder-decoder transformer conditions on text actions and a time-aligned event roll to delete, insert, or enhance individual sound events in dense audio scenes.