Audio Recognition on Speech Commands V2
98.9AccuracyASiT
Evaluation Results
| Method | Links | |
|---|---|---|
| ASiT#Param=86M, Pre-training Data=AS2026.01 | 98.9 | |
| UniRepLKNet-SType=ConvNet, Params=55.5M2023.11 | 98.5 | |
| M2DFine-tuning=true, # Params=86M, Masking ratio=0.62024.04 | 98.5 | |
| A-JEPA#Param=86M, Pre-training Data=AS2026.01 | 98.5 | |
| MSM-MAEFine-tuning=true, # Params=86M, Masking ratio=0.752024.04 | 98.4 | |
| M2DFine-tuning=true, # Params=86M, Masking ratio=0.6, Target input strategy=all patches to target2024.04 | 98.4 | |
| M2DFine-tuning=true, # Params=86M, Masking ratio=0.7, Target input strategy=all patches to target2024.04 | 98.4 | |
| M2DFine-tuning=true, # Params=86M, Masking ratio=0.72024.04 | 98.4 | |
| Audio-MAEPretrain=AS-2M, Type=Transformer, Params=86.2M2023.11 | 98.3 | |
| Audio-MAE (local)Fine-tuning=true, # Params=86M2024.04 | 98.3 | |
| BEATsFine-tuning=true, # Params=90M, Iteration=32024.04 | 98.3 | |
| Audio-MAE#Param=86M, Pre-training Data=AS2026.01 | 98.3 | |
| BEATs_iter3#Param=90M, Pre-training Data=AS2026.01 | 98.3 | |
| EAT#Param=88M, Pre-training Data=AS2026.01 | 98.3 | |
| M2D-CLAP#Param=86M, Pre-training Data=TA+AS2026.01 | 98.3 | |
| ASDA#Param=93M, Pre-training Data=AS2026.01 | 98.3 | |
| CAT#Param=91M, Pre-training Data=AS2026.01 | 98.3 | |
| SSASTFine-tuning=true, # Params=89M2024.04 | 98.2 | |
| EAT-S/MFine-tuning=true, # Params=25.5M2024.04 | 98.15 | |
| ASTFine-tuning=true, # Params=86M, Version=Single2024.04 | 98.11 | |
| BEATsFine-tuning=true, # Params=90M, Iteration=3+2024.04 | 98.1 | |
| AST#Param=86M, Pre-training Data=IN2026.01 | 98.1 | |
| BEATs_iter3+#Param=90M, Pre-training Data=AS2026.01 | 98.1 | |
| ATST-Frame#Param=86M, Pre-training Data=AS2026.01 | 98.1 | |
| MAE-ASTFine-tuning=true, # Params=86M2024.04 | 98 | |
| ATST BaseFine-tuning=true, # Params=86M2024.04 | 98 | |
| HTS-ATFine-tuning=true, # Params=31M2024.04 | 98 | |
| HTS-AT#Param=31M, Pre-training Data=IN2026.01 | 98 | |
| SS-AST#Param=89M, Pre-training Data=AS+LS2026.01 | 98 | |
| ATST-Clip#Param=86M, Pre-training Data=AS2026.01 | 98 | |
| MAE-AST#Param=86M, Pre-training Data=AS+LS2026.01 | 97.9 | |
| SSASTPretrain=AS-2M, Type=Transformer, Params=89.3M2023.11 | 97.8 | |
| MaskSpecFine-tuning=true, # Params=86M2024.04 | 97.7 | |
| MaskSpec#Param=86M, Pre-training Data=AS2026.01 | 97.7 | |
| Meta-TransformerPretrain=LAION-2B, Type=Transformer, Params=86.6M2023.11 | 97 | |
| CLAP#Param=86M, Pre-training Data=TA2026.01 | 96.8 | |
| PSLAPretrain=IN-1K, Type=ConvNet2023.11 | 96.3 | |
| PSLA#Param=14M, Pre-training Data=IN2026.01 | 96.3 | |
| ASTPretrain=AS-2M, Type=Transformer, Params=86.9M2023.11 | 96.2 | |
| DeLoRes-MFine-tuning=true, # Params=5.3M2024.04 | 96 | |
| MSM-MAE#Param=86M, Pre-training Data=AS2026.01 | 87.3 | |
| PANNSType=ConvNet2023.11 | 61.8 | |
| PANN#Param=81M, Pre-training Data=-2026.01 | 61.8 |