Speech Emotion Recognition on MELD-ST Multilingual (test)
68.13W-F1 (SER)ERM-MinMaxGAP
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| ERM-MinMaxGAPModality=Multimodal (Speech + Transcription)2026.03 | 68.13 | 68.85 | 5.08 | 1.74 | 0.38 | 1.24 | 2.11 | |
| ERM-MinMaxGAPModality=Unimodal (Speech only)2026.03 | 58.09 | 60.52 | 8.84 | 2.1 | 0.41 | 0.92 | 3.07 | |
| kimi-audio-7bModality=Unimodal (Speech only)2026.03 | 57.86 | 54.79 | 10.33 | 1.54 | 5.52 | 5.63 | 5.8 | |
| ERM-MinMaxGAPModality=Multimodal (Speech + Transcription)2026.03 | 57.68 | 58.65 | 7.08 | 2.69 | 1.84 | 2.53 | 3.53 | |
| gpt4o-mini-audioModality=Multimodal (Speech + Transcription)2026.03 | 56.99 | 56.35 | 5.77 | 1.36 | 4.42 | 5.29 | 4.21 | |
| kimi-audio-7bModality=Multimodal (Speech + Transcription)2026.03 | 56.11 | 53.85 | 10.58 | 1.71 | 2.02 | 2.17 | 4.77 | |
| Voxtral-Mini-3BModality=Multimodal (Speech + Transcription)2026.03 | 53.74 | 57.71 | 7.64 | 2.04 | 3.57 | 2.46 | 3.93 | |
| gpt4o-mini-audioModality=Multimodal (Speech + Transcription)2026.03 | 52.65 | 51.76 | 8.09 | 2.08 | 4.38 | 5.25 | 4.95 | |
| ERM-MinMaxGAPModality=Unimodal (Speech only)2026.03 | 51.38 | 54.32 | 6.38 | 3.01 | 3.52 | 4.44 | 4.34 | |
| Voxtral-Mini-3BModality=Multimodal (Speech + Transcription)2026.03 | 50.04 | 55.03 | 7.27 | 1.9 | 2.18 | 1.84 | 3.3 | |
| gpt4o-mini-audioModality=Unimodal (Speech only)2026.03 | 49.67 | 49.06 | 12.47 | 1.52 | 4.22 | 5.93 | 6.03 | |
| Qwen2-AudioModality=Multimodal (Speech + Transcription)2026.03 | 46.13 | 41.67 | 10.34 | 2.31 | 4.57 | 5.11 | 5.58 | |
| Qwen2-AudioModality=Unimodal (Speech only)2026.03 | 46.02 | 44.27 | 7.69 | 3.65 | 0.92 | 1.75 | 3.5 | |
| gpt4o-mini-audioModality=Unimodal (Speech only)2026.03 | 45.89 | 44.57 | 9.61 | 1.48 | 2.61 | 4.03 | 4.43 | |
| Voxtral-Mini-3BModality=Unimodal (Speech only)2026.03 | 44.78 | 44.52 | 9.28 | 2.74 | 4.97 | 5.22 | 5.55 | |
| kimi-audio-7bModality=Multimodal (Speech + Transcription)2026.03 | 42.34 | 42.56 | 7.15 | 2.36 | 3.21 | 3.26 | 4 | |
| Voxtral-Mini-3BModality=Unimodal (Speech only)2026.03 | 41.98 | 38.75 | 9.88 | 2.75 | 7.57 | 7.87 | 7.02 | |
| kimi-audio-7bModality=Unimodal (Speech only)2026.03 | 40.27 | 39.96 | 9.18 | 3.12 | 5.21 | 3.58 | 5.27 | |
| Ultravox-0.4Modality=Multimodal (Speech + Transcription)2026.03 | 39.98 | 38.44 | 9.56 | 1.88 | 8.93 | 7.28 | 6.91 | |
| Ultravox-0.4Modality=Unimodal (Speech only)2026.03 | 35.8 | 32.5 | 4.01 | 2.43 | 1.76 | 0.49 | 2.17 | |
| Qwen2-AudioModality=Unimodal (Speech only)2026.03 | 34.89 | 33.31 | 9.78 | 4.26 | 3.36 | 4.67 | 5.51 | |
| Qwen2-AudioModality=Multimodal (Speech + Transcription)2026.03 | 34.62 | 30.79 | 8.66 | 2.08 | 3.67 | 3.35 | 4.44 | |
| Ultravox-0.4Modality=Multimodal (Speech + Transcription)2026.03 | 32.45 | 30.78 | 7.79 | 1.86 | 5.65 | 4.42 | 4.93 | |
| Ultravox-0.4Modality=Unimodal (Speech only)2026.03 | 27.43 | 25.29 | 2.76 | 1.49 | 2.41 | 1.09 | 1.94 |