Composite generation on PlanAudio-Bench
25.2FDPANNsVoiceLDM-s
Evaluation results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| VoiceLDM-sSize=small2026.05 | 25.2 | 379 | 1.39 | 1.53 | 2.71 | 0.15 | 0.7 | 2.35 | |
| VoiceLDM-mSize=medium2026.05 | 22.9 | 363 | 1.32 | 1.41 | 2.86 | 0.19 | 0.09 | 2.81 | |
| AudioLDM2Sound+AudioLDM2Speech2026.05 | 14.3 | 240 | 1.1 | 1.15 | 4.11 | 0.21 | 0.71 | 2.16 | |
| PlanAudio2026.05 | 8.52 | 201 | 0.91 | 1.03 | 3.43 | 0.2 | 0.41 | 2.43 | |
| Reconstruction2026.05 | 3.82 | 112 | 0.23 | 0.37 | 2.82 | 0.21 | 0.21 | 2.49 | |
| GroundTruth2026.05 | 0 | 0 | 0 | 0 | 3.23 | 0.17 | 0.1 | 2.69 |