Audio Tagging on AudioSet balanced (AS-20k)
40.2mAPEAT
Evaluation Results
| Method | Links | |
|---|---|---|
| EAT# Params=88M, Pre-train data=5k2025.10 | 40.2 | |
| SPEARs+a XLarge# Params=600M, Pre-train data=197k2025.10 | 39.4 | |
| SPEARa Large# Params=327M, Pre-train data=5k2025.10 | 39.3 | |
| SPEARa Large# Params=327M, Pre-train data=13k2025.10 | 39.3 | |
| SPEARa Base# Params=94M, Pre-train data=13k2025.10 | 39.2 | |
| SPEARs+a Large# Params=327M, Pre-train data=97k2025.10 | 39.2 | |
| SPEARs+a Base# Params=94M, Pre-train data=97k2025.10 | 39.1 | |
| ATST Frame# Params=86M, Pre-train data=5k2025.10 | 39 | |
| SPEARa Base# Params=94M, Pre-train data=5k2025.10 | 39 | |
| BEATs# Params=90M, Pre-train data=5k2025.10 | 38.9 | |
| USAD Large# Params=330M, Pre-train data=126k, fine-tuned from public checkpoint=true2025.10 | 37.4 | |
| USAD Base# Params=94M, Pre-train data=126k, fine-tuned from public checkpoint=true2025.10 | 35.7 | |
| SPEARs Base# Params=94M, Pre-train data=84k2025.10 | 26.9 | |
| SPEARs Large# Params=327M, Pre-train data=84k2025.10 | 26.4 |