- Who it helps
- Audio researchers and creators following controllable sound editing
- Availability
- November 5 preprint is public. This check did not establish released code, model weights or an available editing service.
Edit a sound rather than describing only its replacement
The paper describes semantic and style changes alongside signal-processing edits, with controls for loudness, edit strength and temporal shape. Its training data is synthesized to pair sounds with instructions.
Listen before drawing a quality conclusion
Quality comparisons are the authors’ reported results. The abstract cannot establish whether an edit preserves the details a particular project needs, and no audio was auditioned here.
What you can do
- Read the primary description and its stated limits before planning an evaluation. Open reference ↗
Where the evidence stops
Abstract checked; no model access, listening test or editing workflow executed.
Sources read. Public source text checked November 14, 2026 UTC. No third-party project, model inference or full workflow executed.
Sources & further reading
Paper · paper
FoleyForge audio editing paper ↗
November 5 abstract describes source-audio editing. No code, weights or audio output verified.
FoleyForge authors · page checked 2026-11-14 · Published 2026-11-05
Coverage note: Abstract checked; no model access, listening test or editing workflow executed.
The record over time (1)
Dated source checks and published explanations. An observation date can differ from the original release date.
- · Added to Current
FoleyForge takes source audio and a text instruction to change sound properties while trying to preserve the original sound’s identity.