Research map / Image generation / Editing and style transfer
AUDIT: Audio Editing by Following Instructions with Latent Diffusion Models
Yuancheng Wang, Zeqian Ju, Xu Tan, Lei He, Zhizheng Wu, Jiang Bian, sheng zhao
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| remove caption text input at inference (ablation) | AUDIT · Inpainting | FD: 18.17 -> 19.03 | Mixed |
| remove caption text input at inference (ablation) | AUDIT · Super-resolution | FD: 18.14 -> 21.97 | Hurt |
| remove caption text input (ablation) | SDEdit-Precise baseline · Inpainting | FD: 21.07 -> 27.63 | Hurt |
| remove caption text input (ablation) | SDEdit-Precise baseline · Super-resolution | FD: 27.81 -> 32.66 | Hurt |
| use precise masked region instead of rough region (ablation) | SDEdit · Inpainting | FD: 21.99 -> 21.07 | Improved |
| replace SDEdit with instruction-trained AUDIT | SDEdit-1/4T · Adding | FD: 28.25 -> 21.80 | Improved |
| replace SDEdit with AUDIT | SDEdit-1/2T · Dropping | FD: 28.19 -> 22.40 | Improved |
| replace SDEdit with AUDIT | SDEdit-1/4T · Replacement | FD: 26.72 -> 21.65 | Improved |
| reduce SDEdit denoising steps from 3/4T to 1/4T | SDEdit · Adding | FD: 28.87 -> 28.25 | Improved |
| reduce SDEdit denoising steps from 3/4T to 1/4T | SDEdit · Dropping | FD: 29.66 -> 31.31 | No clear effect |
| use AUDIT text-to-audio model vs AudioLDM-L-Full | text-to-audio model · AudioCaps | FD: 23.31 -> 20.19 | Improved |
| use AUDIT instead of best baseline (subjective) | AUDIT · Dropping | Relevance: 48.2 -> 81.0 | Improved |
Later papers that compared with this method
1 later accepted paper reports Audit as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Declarative Audio Editing with Audio Language Model (ICLR 2026 · better on 3 of 3 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Prompt-guided Precise Audio Editing with Diffusion Models (ICML 2024 · similarity 0.90)
- SAO-Instruct: Free-form Audio Editing using Natural Language Instructions (NeurIPS 2025 · similarity 0.87)
- Zero-Shot Unsupervised and Text-Based Audio Editing Using DDPM Inversion (ICML 2024 · similarity 0.86)
- AVENUE: Audio-Video Editing Understanding and Evaluation (NeurIPS 2026 · similarity 0.86)
- ZeroSep: Separate Anything in Audio with Zero Training (NeurIPS 2025 · similarity 0.85)
- SoundBrush: Sound as a Brush for Visual Scene Editing (AAAI 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
