Speech Emotion Recognition on IEMOCAP 4-way Full (Session 5 test)
75.4UAR (Variant -)AF3
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| AF3Model Category=Audio language models2026.06 | 75.4 | 77.6 | 2.2 | |
| SmoothCLAPModel Category=Zero-shot contrastive baselines (no tokens)2026.06 | 60.6 | — | — | |
| ParaCLAPModel Category=Zero-shot contrastive baselines (no tokens)2026.06 | 60 | — | — | |
| qw-omniModel Category=Audio language models2026.06 | 54.1 | 58.2 | 4.1 | |
| CLAPModel Category=Zero-shot contrastive baselines (no tokens)2026.06 | 35.3 | — | — | |
| PengiModel Category=Zero-shot contrastive baselines (no tokens)2026.06 | 34.5 | — | — |