Research map / Video / Video generation
AudioX: A Unified Framework for Anything-to-Audio Generation
Zeyue Tian, Zhaoyang Liu, Yizhu Jin, Ruibin Yuan, Liumeng Xue, Xu Tan, Qifeng Chen, Wei Xue, Yike Guo
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 |
|---|---|---|---|
| add Gemini-generated and Qwen-augmented captions (GeminiCap-aug) instead of raw labels (ablation) | AudioX · T2A | FAD: 6.02 -> 2.91 | Improved |
| add Gemini-generated and Qwen-augmented captions instead of raw labels (ablation) | AudioX · instruction-following T2A | Cat-acc: 17.35 -> 28.91 | Improved |
| use GeminiCap-aug instead of AudioSetCaps captions (ablation) | AudioX · T2A | FAD: 3.19 -> 2.91 | Improved |
| use Gemini annotations instead of Qwen2-Audio captions only (ablation) | AudioX · T2A | FAD: 4.40 -> 3.02 | Improved |
| add Qwen2-Audio augmentation on top of Gemini captions (ablation) | AudioX · T2A | FAD: 3.02 -> 2.91 | Improved |
| add Qwen2-Audio augmentation on top of Gemini captions (ablation) | AudioX · V2A | FAD: 1.31 -> 1.15 | Improved |
| better text captions during training transfer to video-to-audio (ablation) | AudioX · V2A | FAD: 1.81 -> 1.15 | Improved |
| add MAF module (ablation) | AudioX · AudioCaps/T2A ablation | FAD: 2.67 -> 1.98 | Improved |
| add gate to MAF (ablation) | AudioX · T2A ablation | FAD: 2.00 -> 1.98 | Improved |
| add query-based attention to MAF (ablation) | AudioX · T2A ablation | FAD: 2.08 -> 1.98 | Improved |
| add gate to MAF (ablation) | AudioX · T2A ablation | Ordering: 0.876 -> 0.888 | Hurt |
| add GeminiCap-aug captions (ablation) | AudioX · TV2A | FAD: 3.41 -> 1.48 | Improved |
Later papers that compared with this method
3 later accepted papers report AudioX as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- Omni2Sound: Towards Unified Video-Text-to-Audio Generation (CVPR 2026 · better on 5 of 5 shared results)
- AV-Edit: Multimodal Generative Sound Effect Editing via Audio-Visual Semantic Joint Control (AAAI 2026 · better on 4 of 4 shared results)
- SALSA-V: Shortcut-Augmented Long-form Synchronized Audio from Videos (ICML 2026 · better on 2 of 3 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- VidMuse — VidMuse: A Simple Video-to-Music Generation Framework with Long-Short-Term Modeling (CVPR 2025) · better on 1 of 1 shared result
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
- Read, Watch and Scream! Sound Generation from Text and Video (AAAI 2025 · similarity 0.90)
- UniFlow-Audio: Unified Flow Matching for Audio Generation from Omni-Modalities (NeurIPS 2026 · similarity 0.88)
- Video-Guided Foley Sound Generation with Multimodal Controls (CVPR 2025 · similarity 0.88)
- VinTAGe: Joint Video and Text Conditioning for Holistic Audio Generation (CVPR 2025 · similarity 0.88)
- Audio-Sync Video Generation with Multi-Stream Temporal Control (NeurIPS 2025 · similarity 0.88)
- AudioGenX: Explainability on Text-to-Audio Generative Models (AAAI 2025 · similarity 0.88)
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.
