Speech Emotion Recognition on EmoDB Zero-Shot v1 (test)
87.85AccuracyQwen2.5-Omni
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Qwen2.5-OmniSemantic Modality=true, Acoustic Modality=false2026.01 | 87.85 | 85.97 | |
| Qwen2-AudioSemantic Modality=true, Acoustic Modality=false2026.01 | 74.21 | 70.29 | |
| Emotion2VecSemantic Modality=false, Acoustic Modality=false2026.01 | 71.21 | 76.07 | |
| WhisperSemantic Modality=true, Acoustic Modality=false2026.01 | 68.84 | 63.46 | |
| FASSemantic Modality=true, Acoustic Modality=true2026.01 | 68.1 | 65.07 | |
| WavLMSemantic Modality=true, Acoustic Modality=true2026.01 | 64.18 | 56.74 | |
| HuBERTSemantic Modality=true, Acoustic Modality=true2026.01 | 52.99 | 53.57 | |
| CLAPSemantic Modality=true, Acoustic Modality=false2026.01 | 43.28 | 41.45 | |
| MingTok-AudioSemantic Modality=false, Acoustic Modality=true2026.01 | 37.31 | 34.02 | |
| EnCodecSemantic Modality=false, Acoustic Modality=true2026.01 | 29.1 | 23.87 | |
| VibevoiceSemantic Modality=false, Acoustic Modality=true2026.01 | 28.73 | 20.92 | |
| wav2vec 2.0Semantic Modality=true, Acoustic Modality=true2026.01 | 25.37 | 22.09 |