Speech Emotion Recognition on SER (nAUC Budgeted Evaluation)
7.6nAUC (0-5 pp budget)GACL
Evaluation results
| Method | Links | ||
|---|---|---|---|
| GACLModel=Qwen2.5-Omni2026.06 | 7.6 | 11.4 | |
| GACLModel=Voxtral-Small2026.06 | 6.7 | 10.3 | |
| GACLModel=Qwen3-Omni2026.06 | 6.6 | 10.2 | |
| ACDModel=Qwen2.5-Omni2026.06 | 3.9 | 5.7 | |
| AADModel=Qwen2.5-Omni2026.06 | 3.4 | 6.8 | |
| GACLModel=Qwen2-Audio-Inst.2026.06 | 2.9 | 6.1 | |
| AADModel=Voxtral-Small2026.06 | 2.8 | 4.4 | |
| AADModel=Qwen2-Audio-Inst.2026.06 | 2.3 | 3.3 | |
| ACDModel=Voxtral-Small2026.06 | 2.3 | 2.9 | |
| GACLModel=Kimi-Audio-Inst.2026.06 | 2 | 4.9 | |
| ACDModel=Qwen3-Omni2026.06 | 0.8 | 1.2 | |
| ACDModel=Kimi-Audio-Inst.2026.06 | 0.8 | 1.1 | |
| AADModel=Kimi-Audio-Inst.2026.06 | 0.5 | 0.7 | |
| ACDModel=Qwen2-Audio-Inst.2026.06 | 0 | 0 | |
| AADModel=Qwen3-Omni2026.06 | 0 | 0 |