CAV-MAE (Audio-Only)

Reported on 5 benchmarks across 3 tasks · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Audio3 results

Audio ClassificationonAudioSet
Test mAP· uses extra data· 2022-10-02
0.466
best: 0.558 (OmniVec2)
Contrastive Audio-Visual Masked Autoencoder arXiv:2210.07839
Audio ClassificationonVGGSound
Top 1 Accuracy· uses extra data· 2022-10-02
59.5
best: 69.8 (Mirasol3B)
Contrastive Audio-Visual Masked Autoencoder arXiv:2210.07839
Audio TaggingonAudioSet
mean average precision· uses extra data· 2022-10-02
0.466
best: 0.512 (CAV-MAE (Audio-Visual))
Contrastive Audio-Visual Masked Autoencoder arXiv:2210.07839

Methodology2 results

ClassificationonAudioSet
Test mAP· uses extra data· 2022-10-02
0.466
best: 0.558 (OmniVec2)
Contrastive Audio-Visual Masked Autoencoder arXiv:2210.07839
ClassificationonVGGSound
Top 1 Accuracy· uses extra data· 2022-10-02
59.5
best: 69.8 (Mirasol3B)
Contrastive Audio-Visual Masked Autoencoder arXiv:2210.07839