Visually Guided Mono-to-Binaural Audio Generation on FAIR-Play (10-split)
0.82STFT ErrorSIREN
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| SIREN2026.03 | 0.82 | 0.132 | 1.55 | 7.219 | |
| CC-Stereo2026.03 | 0.823 | 0.132 | — | 7.144 | |
| CMC2026.03 | 0.849 | 0.133 | 1.423 | — | |
| Sep-Stereo2026.03 | 0.879 | 0.136 | 1.429 | — | |
| Mono2Binaural2026.03 | 0.889 | 0.137 | 1.438 | 6.232 |