ML Research Wiki / Benchmarks / Audio Classification / FSD50K

FSD50K

Audio Classification Benchmark

Performance Over Time

📊 Showing 10 results | 📏 Metric: mAP

Top Performing Models

Rank Model Paper mAP Date Code
1 ONE-PEACE 📚 ONE-PEACE: Exploring One General Representation Model Toward Unlimited Modalities 69.70 2023-05-18 📦 modelscope/modelscope 📦 OFA-Sys/ONE-PEACE
2 MN 📚 Dynamic Convolutional Neural Networks as Efficient Pre-trained Audio Models 65.60 2023-10-24 📦 fschmid56/efficientat
3 PaSST-S 📚 Efficient Training of Audio Transformers with Patchout 65.55 2021-10-11 📦 kkoutini/passt 📦 kkoutini/passt_hear21
4 DyMN-L 📚 Dynamic Convolutional Neural Networks as Efficient Pre-trained Audio Models 65.50 2023-10-24 📦 fschmid56/efficientat
5 PaSST-N-S 📚 Efficient Training of Audio Transformers with Patchout 64.20 2021-10-11 📦 kkoutini/passt 📦 kkoutini/passt_hear21
6 LHGNN LHGNN: Local-Higher Order Graph Neural Networks For Audio Classification and Tagging 59.00 2025-01-07 -
7 PSLA 📚 PSLA: Improving Audio Tagging with Pretraining, Sampling, Labeling, and Aggregation 56.71 2021-02-02 📦 YuanGongND/psla
8 MATPAC (SSL Model) Masked Latent Prediction and Classification for Self-Supervised Audio Representation Learning 55.20 2025-02-17 📦 aurianworld/matpac
9 Temporal Knowledge Distillation for On-device Audio Classification Temporal Knowledge Distillation for On-device Audio Classification 54.80 2021-10-27 -
10 Large 6-Layer Transformer with Pooling Audio Transformers 53.70 2021-05-01 -

All Papers (10)

modelscope/modelscope OFA-Sys/ONE-PEACE
fschmid56/efficientat
kkoutini/passt kkoutini/passt_hear21
fschmid56/efficientat
kkoutini/passt kkoutini/passt_hear21
YuanGongND/psla
aurianworld/matpac

Temporal Knowledge Distillation for On-device Audio Classification

2021
Temporal Knowledge Distillation for On-device Audio Classification

Audio Transformers

2021
Large 6-Layer Transformer with Pooling