SepTr + LeRaC

Reported on 6 benchmarks across 4 tasks · 1 paper

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

Audio3 results

Emotion RecognitiononCREMA-D
Accuracy· 2022-05-18
70.95
best: 94.07 (Vertically long patch ViT)
Learning Rate Curriculum arXiv:2205.09180
Audio ClassificationonESC-50
Accuracy (5-fold)· 2022-05-18
91.58
best: 99.1 (OmniVec2)
Learning Rate Curriculum arXiv:2205.09180
Audio ClassificationonESC-50
Top-1 Accuracy· 2022-05-18
91.58
best: 99.1 (OmniVec2)
Learning Rate Curriculum arXiv:2205.09180

Methodology2 results

ClassificationonESC-50
Accuracy (5-fold)· 2022-05-18
91.58
best: 99.1 (OmniVec2)
Learning Rate Curriculum arXiv:2205.09180
ClassificationonESC-50
Top-1 Accuracy· 2022-05-18
91.58
best: 99.1 (OmniVec2)
Learning Rate Curriculum arXiv:2205.09180

Speech1 result

Speech Emotion RecognitiononCREMA-D
Accuracy· 2022-05-18
70.95
best: 94.07 (Vertically long patch ViT)
Learning Rate Curriculum arXiv:2205.09180