Emotion Recognition on MSP-Podcast
69.4AccuracyCCMT
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| CCMTAudio=✓, Text (En)=✓, Text (Es)=✗, Text (De)=✗, Text (Fr)=✓2026.07 | 69.4 | — | — | — | — | — | 68.26 | 25,082 | |
| CCMTAudio=✓, Text (En)=✓, Text (Es)=✗, Text (De)=✓, Text (Fr)=✗2026.07 | 69.24 | — | — | — | — | — | 68.28 | 27,714 | |
| CCMTAudio=✓, Text (En)=✓, Text (Es)=✗, Text (De)=✗, Text (Fr)=✗2026.07 | 69.1 | — | — | — | — | — | 68.12 | 3,404 | |
| CCMTAudio=✓, Text (En)=✓, Text (Es)=✓, Text (De)=✓, Text (Fr)=✓2026.07 | 69.03 | — | — | — | — | — | 68.12 | 76,000 | |
| CCMTAudio=✓, Text (En)=✓, Text (Es)=✓, Text (De)=✗, Text (Fr)=✗2026.07 | 68.81 | — | — | — | — | — | 67.93 | 30,112 | |
| WavLM (KD from Audio + En + Fr CCMT)Audio=✓, Text (En)=✗, Text (Es)=✗, Text (De)=✗, Text (Fr)=✗, Distillation Source=Audio + En + Fr CCMT2026.07 | 65.06 | — | — | — | — | — | 63.93 | 211 | |
| WavLM (KD from Audio + En + De CCMT)Audio=✓, Text (En)=✗, Text (Es)=✗, Text (De)=✗, Text (Fr)=✗, Distillation Source=Audio + En + De CCMT2026.07 | 64.74 | — | — | — | — | — | 63.29 | 210 | |
| WavLMAudio=✓, Text (En)=✗, Text (Es)=✗, Text (De)=✗, Text (Fr)=✗2026.07 | 64.25 | — | — | — | — | — | 62.39 | 213 | |
| RoBERTaAudio=✗, Text (En)=✓, Text (Es)=✗, Text (De)=✗, Text (Fr)=✗2026.07 | 62.35 | — | — | — | — | — | 61.02 | 27 | |
| RoBERTuitoAudio=✗, Text (En)=✗, Text (Es)=✓, Text (De)=✗, Text (Fr)=✗2026.07 | 59.81 | — | — | — | — | — | 57.51 | 138 | |
| GBERTAudio=✗, Text (En)=✗, Text (Es)=✗, Text (De)=✓, Text (Fr)=✗2026.07 | 58.93 | — | — | — | — | — | 57.42 | 139 | |
| CamemBERTAudio=✗, Text (En)=✗, Text (Es)=✗, Text (De)=✗, Text (Fr)=✓2026.07 | 58.67 | — | — | — | — | — | 57.06 | 138 | |
| WhisperAudio=✓, Text (En)=✗, Text (Es)=✗, Text (De)=✗, Text (Fr)=✗2026.07 | 57.23 | — | — | — | — | — | 56.99 | 180 | |
| AER-LLMEvaluation Setting=Few-Shot (FS)2026.01 | 55.6 | 0.4 | 0.61 | 0.56 | — | 56.2 | — | — | |
| AmbER22026.01 | 52 | 0.33 | 0.71 | 0.43 | 0.369 | 44.5 | — | — | |
| AER-LLMEvaluation Setting=Zero-Shot (ZS)2026.01 | 50.6 | 0.45 | 0.54 | 0.52 | — | 50.5 | — | — | |
| SHALA-LLM2026.06 | 49.6 | 0.544 | 0.694 | — | 30.1 | 45.5 | — | — | |
| MLS2026.06 | 48.8 | 0.58 | 0.585 | — | 23.3 | 41.5 | — | — | |
| SHALA-LLM (w/o Ambi-En)Ambi-En=false2026.06 | 48.2 | 0.55 | 0.658 | — | 27.6 | 43.3 | — | — | |
| Zero-shotinference_mode=Zero-shot2026.06 | 42.1 | 0.64 | 0.508 | — | 26.6 | 38.8 | — | — | |
| EMO-Super2026.01 | — | — | — | — | 0.363 | — | — | — |