SepTr

Reported on 5 benchmarks across 5 tasks · 1 paper · 1 SOTA

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

Audio2 results

Emotion RecognitiononCREMA-D
Accuracy· 2022-03-17
70.47
best: 94.07 (Vertically long patch ViT)
SepTr: Separable Transformer for Audio Spectrogram Processing arXiv:2203.09581
Audio ClassificationonESC-50
Top-1 Accuracy· 2022-03-17
91.13
best: 99.1 (OmniVec2)
SepTr: Separable Transformer for Audio Spectrogram Processing arXiv:2203.09581

Time Series1 result

Time Series AnalysisonSpeech Commands
% Test Accuracy· 2022-03-17
98.51
SOTA
SepTr: Separable Transformer for Audio Spectrogram Processing arXiv:2203.09581

Methodology1 result

ClassificationonESC-50
Top-1 Accuracy· 2022-03-17
91.13
best: 99.1 (OmniVec2)
SepTr: Separable Transformer for Audio Spectrogram Processing arXiv:2203.09581

Speech1 result

Speech Emotion RecognitiononCREMA-D
Accuracy· 2022-03-17
70.47
best: 94.07 (Vertically long patch ViT)
SepTr: Separable Transformer for Audio Spectrogram Processing arXiv:2203.09581