Binaural Audio Generation on FAIR-Play
1.155STFT ScoreMono-Mono
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Mono-MonoGuidance=None (Duplicating mono audio)2025.06 | 1.155 | 0.153 | 7.666 | 5.735 | |
| L2BNetGuidance=Visually-guided2025.06 | 1.028 | 0.148 | — | — | |
| Mono2BinauralGuidance=Visually-guided2025.06 | 0.959 | 0.141 | 6.496 | 6.232 | |
| TASGuidance=Text-guided2025.06 | 0.914 | 0.137 | 6.092 | 6.771 | |
| APNetGuidance=Visually-guided2025.06 | 0.889 | 0.136 | 5.758 | 6.972 | |
| Sep-binauralGuidance=Visually-guided2025.06 | 0.879 | 0.135 | 6.526 | 6.422 | |
| Main-netGuidance=Visually-guided2025.06 | 0.867 | 0.135 | 5.75 | 6.985 | |
| Complete-netGuidance=Visually-guided2025.06 | 0.856 | 0.134 | 5.787 | 6.959 | |
| AVSNGuidance=Visually-guided2025.06 | 0.849 | 0.133 | — | — | |
| CyclicGuidance=Visually-guided2025.06 | 0.787 | 0.128 | 5.244 | 7.546 | |
| Text-guided Audio SpatializationGuidance=Text-guided2025.06 | 0.773 | 0.126 | 5.019 | 7.966 |