Audio Classification on AudioSet
49.6mAPAbsolute SOTA
Evaluation Results
| Method | Links | |
|---|---|---|
| Absolute SOTAAnchor Modality=–2026.04 | 49.6 | |
| ASTModel Architecture=Pure Attention, Ensemble Config=Ensemble-M2021.04 | 48.5 | |
| ASTModel Architecture=Pure Attention, Ensemble Config=Ensemble-S2021.04 | 47.5 | |
| PSLAModel Architecture=CNN+Attention, Ensemble Config=Ensemble-M2021.04 | 47.4 | |
| AV-MAEPretraining=SSL AS-2M, Training set=AS-500K2022.12 | 46.6 | |
| PSLAModel Architecture=CNN+Attention, Ensemble Config=Ensemble-S2021.04 | 46.4 | |
| ASTModel Architecture=Pure Attention, Ensemble Config=Single2021.04 | 45.9 | |
| PSLAModel Architecture=CNN+Attention, Ensemble Config=Single2021.04 | 44.4 | |
| SupervisedProtocol=MLP2020.06 | 43.9 | |
| PANNsModel Architecture=CNN+Attention2021.04 | 43.9 | |
| MBTPretraining=Im21K, Training set=AS-500K2022.12 | 41.5 | |
| PerceiverPretraining=None, Training set=AS-2M2022.12 | 38.4 | |
| Fayek et al.Pretraining=Im1K, Training set=AS-2M2022.12 | 38.4 | |
| GBlendPretraining=Im1K, Training set=AS-2M2022.12 | 32.4 | |
| BaselineModel Architecture=CNN+MLP2021.04 | 31.4 | |
| MMV FACBackbone=TSM-50x2, Train data=AS+HT2020.06 | 30.9 | |
| VA onlyBackbone=S3D-G, Train data=AS+HT2020.06 | 30.6 | |
| MMV FACBackbone=TSM-50, Train data=AS+HT2020.06 | 30.6 | |
| MMV FACBackbone=S3D-G, Train data=AS+HT2020.06 | 30.3 | |
| LanguageBind2026.05 | 30 | |
| VA onlyBackbone=R(2+1)D-18, Train data=AS2020.06 | 29.7 | |
| VA onlyBackbone=S3D-G, Train data=AS2020.06 | 29.7 | |
| CodeBind-VL2026.05 | 29.2 | |
| AA+AV CCBackbone=RN-50 (23.5M), Train data=AS, Modality=VA2020.06 | 28.5 | |
| AudioCLIPAnchor Modality=–2026.04 | 28.4 | |
| EmergentBridgeAnchor Modality=Text2026.04 | 28.1 | |
| LanguageBindAnchor Modality=Text2026.04 | 27.7 | |
| ViT-LENS_Lanchor=I+T2023.11 | 26.7 | |
| AudioCLIPanchor=I+T2023.11 | 25.9 | |
| EmergentBridgeAnchor Modality=Image2026.04 | 25.4 | |
| OmniBind2026.05 | 25.1 | |
| UNIALIGN2026.05 | 23.8 | |
| ViT-LENS_Lanchor=I2023.11 | 23.1 | |
| CLAPAnchor Modality=–2026.04 | 23.1 | |
| SAM-780Mr=256, Conn.=(a), Experiment=E52025.09 | 21.1 | |
| SAM-2.7Br=256, Conn.=(a), Experiment=E62025.09 | 21.1 | |
| SAM-2.7Br=256, Conn.=(b), Ablation=4.2, Experiment=E142025.09 | 21.1 | |
| CodeBind-IB2026.05 | 21.1 | |
| SAM-2.7Br=256, Conn.=(a), Ablation=4.1, Experiment=E92025.09 | 21 | |
| SAM-2.7Br=8, Conn.=(a), Experiment=E32025.09 | 20.8 | |
| SAM-2.7Br=256, Conn.=(a), Ablation=4.3, Experiment=E182025.09 | 20.8 | |
| SAM-2.7Br=256, Conn.=(c), Ablation=4.2, Experiment=E152025.09 | 20.7 | |
| SAM-780Mr=256, Conn.=(a), Ablation=4.3, Experiment=E172025.09 | 20.7 | |
| SAM-780Mr=256, Conn.=(c), Ablation=4.2, Experiment=E132025.09 | 20.5 | |
| SAM-780Mr=256, Conn.=(a), Ablation=4.1, Experiment=E82025.09 | 20.4 | |
| SAM-780Mr=8, Conn.=(a), Experiment=E22025.09 | 20.3 | |
| SAM-780Mr=256, Conn.=(b), Ablation=4.2, Experiment=E122025.09 | 20 | |
| SAM-130Mr=256, Conn.=(a), Experiment=E42025.09 | 19.9 | |
| SAM-130Mr=8, Conn.=(a), Experiment=E12025.09 | 19.7 | |
| FreeBind2026.05 | 19.7 | |
| ssLALM-2.8Br=8, Conn.=Linear2025.09 | 19.4 | |
| GAMA-7Br=8, Conn.=QFormer2025.09 | 19.2 | |
| SAM-130Mr=256, Conn.=(c), Ablation=4.2, Experiment=E112025.09 | 19.1 | |
| SAM-130Mr=256, Conn.=(a), Ablation=4.3, Experiment=E162025.09 | 19 | |
| SAM-130Mr=256, Conn.=(b), Ablation=4.2, Experiment=E102025.09 | 18.8 | |
| LTU-7Br=8, Conn.=Linear2025.09 | 18.7 | |
| SAM-130Mr=256, Conn.=(a), Ablation=4.1, Experiment=E72025.09 | 18.2 | |
| ImageBind-Hanchor=I2023.11 | 17.6 | |
| ImageBindAnchor Modality=Image2026.04 | 17.6 | |
| WAV2CLIPAnchor Modality=–2026.04 | 0.71 |