Joint Video-Audio Generation on Open-domain (test)
0.096AV-align_bindDiffusion Latent Aligner
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| Diffusion Latent Alignervariant=full, domain=Open-domain, denoising_steps=252024.02 | 0.096 | 0.324 | 0.138 | 0.283 | |
| Diffusion Latent Alignervariant=vanilla, domain=Open-domain, denoising_steps=252024.02 | 0.074 | 0.322 | 0.081 | 0.226 |