Audio Classification on AudioSet

102개 결과 · ⬇ CSV · JSON

Test mAP

Rank	Model	Test mAP	AUC	d-prime	Extra Training Data	Paper	Code	Year
1	OmniVec2	0.558	–	–	✓	OmniVec2 - A Novel Transformer based Network for Large Scale Multimodal and Multitask Learning		2024
2	OmniVec	0.548	–	–	✓	OmniVec: Learning robust representations with cross modal sharing		2023
3	EquiAV	0.546	–	–		EquiAV: Leveraging Equivariance for Audio-Visual Contrastive Learning	jongsuk1/equiav	2024
4	MAViL (Audio-Visual, single)	0.533	–	–	✓
5	Audiovisual Masked Autoencoder (Audiovisual, Single)	0.518	–	–		Audiovisual Masked Autoencoders	google-research/scenic · google-research/scenic	2022
6	CAV-MAE (Audio-Visual)	0.512	–	–	✓	Contrastive Audio-Visual Masked Autoencoder	yuangongnd/cav-mae	2022
7	BEATs (Audio-only, Ensemble)	0.506	–	–		BEATs: Audio Pre-Training with Acoustic Tokenizers	microsoft/unilm · Yui010206/CREMA · qingyuliu0521/icsd · +1	2022
8	UAVM (Audio + Video)	0.504	–	–	✓	UAVM: Towards Unifying Audio and Visual Models	YuanGongND/uavm	2022
9	SSLAM (Audio-Only, Single)	0.502	–	–		SSLAM: Enhancing Self-Supervised Models with Audio Mixtures for Polyphonic Soundscapes	ta012/SSLAM	2025
10	mn40_as (Ensemble)	0.498	–	–	✓	Efficient Large-scale Audio Tagging via Transformer-to-CNN Knowledge Distillation	fschmid56/efficientat · fschmid56/efficientat_hear	2022
11	ATST-C2F(Single)	0.497	–	–		Self-supervised Audio Teacher-Student Transformer for Both Clip-level and Frame-level Tasks	Audio-WestlakeU/ATST-SED · audio-westlakeu/audiossl	2023
12	MBT (AS-500K training + Video)	0.496	–	–	✓	Attention Bottlenecks for Multimodal Fusion	google-research/scenic	2021
12	PaSST (Ensemble)	0.496	–	–	✓	Efficient Training of Audio Transformers with Patchout	kkoutini/passt · kkoutini/passt_hear21	2021
14	DyMN-L (Audio-Only, Single)	0.490	–	–	✓	Dynamic Convolutional Neural Networks as Efficient Pre-trained Audio Models	fschmid56/efficientat	2023
14	M2D2	0.490	–	–		M2D2: Exploring General-purpose Audio-Language Representations Beyond CLAP	nttcslab/m2d · nttcslab/eval-audio-repr	2025
16	HTS-AT (Ensemble)	0.487	–	–	✓	HTS-AT: A Hierarchical Token-Semantic Audio Transformer for Sound Classification and Detection	retrocirce/hts-audio-transformer	2022
17	BEATs (Audio-only, Single)	0.486	–	–		BEATs: Audio Pre-Training with Acoustic Tokenizers	microsoft/unilm · Yui010206/CREMA · qingyuliu0521/icsd · +1	2022
17	EAT	0.486	–	–		EAT: Self-Supervised Pre-Training with Efficient Audio Transformer	cwx-worst-one/eat	2024
17	DTF-AT (Single)	0.486	–	–		DTF-AT: Decoupled Time-Frequency Audio Transformer for Event Classification	ta012/DTFAT	2024
20	AST (Ensemble)	0.485	–	–	✓	AST: Audio Spectrogram Transformer	YuanGongND/ast · nttcslab/composing-general-audio-repr · pxaris/ccml · +2	2021