Audio Reconstruction on SongDescriber no vocals 10 s chunks (eval)
1.27STFT ErrorSAO Baseline
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| SAO Baselinetype=Baseline2025.07 | 1.27 | 0.77 | 6.5 | |
| T5 aligned Re-Bottleneckalignment_target=T52025.07 | 1.29 | 0.79 | 6.2 | |
| BEATs aligned Re-Bottleneckalignment_target=BEATs2025.07 | 1.29 | 0.81 | 6.2 | |
| Equivariant Re-Bottleneckproperty=Equivariant2025.07 | 1.39 | 0.87 | 6.3 |