Image-to-Audio generation on VGGSound Paprika style (test)
2.612KL DivergenceIm2Wav
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Im2Wav2024.02 | 2.612 | 7.055 | 19.627 | 7.576 | |
| Diffusion Latent Alignervariant=full, denoising_steps=302024.02 | 2.691 | 6.149 | 20.958 | 6.869 | |
| Diffusion Latent Alignervariant=vanilla, denoising_steps=302024.02 | 3.115 | 4.986 | 33.049 | 7.364 |