Audio Event Addition on Event-level Editing Benchmark
47.74CLAP ScoreSDEdit
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| SDEditBackbone=Tango22026.06 | 47.74 | 4.049 | 1.02 | 5.886 | 0.354 | |
| DIRECTAUDIOEDITBackbone=Tango22026.06 | 47.65 | 3.103 | 0.674 | 6.709 | 0.655 | |
| DDPM-InvBackbone=Tango22026.06 | 47.28 | 3.5 | 0.796 | 6.239 | 0.634 | |
| SDEditBackbone=AudioLDM22026.06 | 43.03 | 2.166 | 1.257 | 6.626 | 0.311 | |
| DDPM-InvBackbone=AudioLDM22026.06 | 42.23 | 1.046 | 0.721 | 6.857 | 0.688 | |
| DIRECTAUDIOEDITBackbone=AudioLDM22026.06 | 41.44 | 0.989 | 0.536 | 6.709 | 0.702 | |
| DDIM-InvBackbone=Tango22026.06 | 39.38 | 6.547 | 2.094 | 5.251 | 0.479 | |
| DDIM-InvBackbone=AudioLDM22026.06 | 34.16 | 3.428 | 2.154 | 6.073 | 0.535 |