ATST-SED

Reported on 3 benchmarks across 1 task · 1 paper · 2 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Audio3 results

Sound Event DetectiononDESED
PSDS1· uses extra data· 2023-09-15
0.583
best: 0.587 (MAT-SED)
SOTA
Fine-tune the pretrained ATST model for sound event detection arXiv:2309.08153
Sound Event DetectiononDESED
event-based F1 score· uses extra data· 2023-09-15
63.4
SOTA
Fine-tune the pretrained ATST model for sound event detection arXiv:2309.08153
Sound Event DetectiononDESED
PSDS2· uses extra data· 2023-09-15
0.81
best: 0.896 (MAT-SED)
Fine-tune the pretrained ATST model for sound event detection arXiv:2309.08153