Audio Captioning on AudioCaps (MLLM/Human Evaluation)
0.75MWR-S (MLLM)SoundAtlas
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| SoundAtlas2026.01 | 0.75 | 0.58 | 0.71 | 0.69 | |
| Auto-ACD2026.01 | 0.39 | 0.41 | 0.31 | 0.26 | |
| Human-Expert2026.01 | 0.36 | 0.51 | 0.46 | 0.55 |
| Method | Links | ||||
|---|---|---|---|---|---|
| SoundAtlas2026.01 | 0.75 | 0.58 | 0.71 | 0.69 | |
| Auto-ACD2026.01 | 0.39 | 0.41 | 0.31 | 0.26 | |
| Human-Expert2026.01 | 0.36 | 0.51 | 0.46 | 0.55 |