Audio Generation on AudioCaps
Metric: FD_openl3 (higher is better)
Metric: FD_openl3 (higher is better)
| # | Model↕ | FD_openl3▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | AudioGen | 185.53 | No | AudioGen: Textually Guided Audio Generation | 2022-09-30 | Code |
| 2 | AudioLDM2-large | 158.04 | No | AudioLDM 2: Learning Holistic Audio Generation w... | 2023-08-10 | Code |
| 3 | Stable Audio 2.0 | 110.62 | No | Long-form music generation with latent diffusion | 2024-04-16 | Code |
| 4 | Stable Audio | 103.66 | No | Fast Timing-Conditioned Latent Audio Diffusion | 2024-02-07 | Code |
| 5 | ETTA | 80.13 | No | ETTA: Elucidating the Design Space of Text-to-Au... | 2024-12-26 | Code |
| 6 | TangoFlux-base | 79.7 | No | TangoFlux: Super Fast and Faithful Text to Audio... | 2024-12-30 | Code |
| 7 | Stable Audio Open | 78.24 | No | Stable Audio Open | 2024-07-19 | Code |
| 8 | TangoFlux | 75.1 | No | TangoFlux: Super Fast and Faithful Text to Audio... | 2024-12-30 | Code |
| 9 | ETTA-FT-AC-100k | 61.79 | No | ETTA: Elucidating the Design Space of Text-to-Au... | 2024-12-26 | Code |