SACL-LSTM (one seed)

Reported on 6 benchmarks across 1 task · 1 paper · 1 SOTA

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

Audio6 results

Emotion RecognitiononMELD
Accuracy· 2023-06-02
67.89
best: 68.7 (ELR-GNN)
SOTA
Supervised Adversarial Contrastive Learning for Emotion Recognition in Conversations arXiv:2306.01505
Emotion RecognitiononEmoryNLP
Micro-F1· 2023-06-02
43.19
best: 44.31 (EmotionIC)
Supervised Adversarial Contrastive Learning for Emotion Recognition in Conversations arXiv:2306.01505
Emotion RecognitiononEmoryNLP
Weighted-F1· 2023-06-02
40.47
best: 42.08 (CKERC)
Supervised Adversarial Contrastive Learning for Emotion Recognition in Conversations arXiv:2306.01505
Emotion RecognitiononMELD
Weighted-F1· 2023-06-02
66.86
best: 69.9 (ELR-GNN)
Supervised Adversarial Contrastive Learning for Emotion Recognition in Conversations arXiv:2306.01505
Emotion RecognitiononIEMOCAP
Accuracy· 2023-06-02
69.62
best: 73.95 (SDT)
Supervised Adversarial Contrastive Learning for Emotion Recognition in Conversations arXiv:2306.01505
Emotion RecognitiononIEMOCAP
Weighted-F1· 2023-06-02
69.7
best: 74.08 (SDT)
Supervised Adversarial Contrastive Learning for Emotion Recognition in Conversations arXiv:2306.01505