Alignment Detection on AnyAudio-Judge English
91.42Accuracy (Speech, Real)AnyAudio-Judge
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| AnyAudio-JudgePrompt=dynamic rubric2026.06 | 91.42 | 78.85 | 78.7 | 84.67 | 91.25 | 79.87 | 86.4 | 84.45 | |
| Gemini-2.5-ProPrompt=holistic2026.06 | 91.22 | 79.2 | 70.23 | 72.6 | 83.25 | 78.87 | 69.8 | 77.72 | |
| Gemini-2.5-ProPrompt=dynamic rubric2026.06 | 89 | 77.67 | 71.67 | 76 | 79.33 | 76.19 | 71 | 77.27 | |
| Qwen3-Omni-30B-A3B-InstructPrompt=dynamic rubric2026.06 | 88.92 | 77.85 | 73.3 | 78.92 | 78.89 | 77.38 | 66.1 | 77.34 | |
| Qwen3-Omni-30B-A3B-CaptionerPrompt=dynamic rubric2026.06 | 88.5 | 77.35 | 72.1 | 78.42 | 78.33 | 76 | 66.7 | 76.77 | |
| Kimi-Audio-7B-InstructPrompt=dynamic rubric2026.06 | 80.08 | 69.15 | 68 | 74 | 70.42 | 68.25 | 65.8 | 70.81 | |
| Qwen2.5-Omni-7BPrompt=dynamic rubric2026.06 | 77.25 | 69.9 | 67.4 | 76.25 | 77.08 | 71.37 | 66.4 | 72.24 | |
| Qwen3-Omni-30B-A3B-CaptionerPrompt=holistic2026.06 | 70.46 | 61.7 | 60.4 | 66.58 | 67.92 | 70.75 | 51.9 | 64.24 | |
| Qwen3-Omni-30B-A3B-InstructPrompt=holistic2026.06 | 65.42 | 58.45 | 54.1 | 58.75 | 68.61 | 70.25 | 52.5 | 61.15 | |
| MiDashengLMPrompt=dynamic rubric2026.06 | 64.92 | 63.6 | 69 | 74.75 | 72.78 | 70.63 | 59.9 | 67.94 | |
| Audio-Flamingo3Prompt=dynamic rubric2026.06 | 59 | 58.4 | 67.6 | 68.42 | 70.97 | 67.75 | 57.2 | 64.19 | |
| Kimi-Audio-7B-InstructPrompt=holistic2026.06 | 50.25 | 50.15 | 51.3 | 49.58 | 50.83 | 50.38 | 48.6 | 50.16 | |
| Qwen2.5-Omni-7BPrompt=holistic2026.06 | 50.25 | 50.2 | 50.4 | 51.25 | 56.25 | 53.88 | 50.1 | 51.76 | |
| Audio-Flamingo3Prompt=holistic2026.06 | 48.87 | 51.26 | 63.78 | 67.58 | 69.32 | 64.78 | 51.65 | 59.61 | |
| MiDashengLMPrompt=holistic2026.06 | 48.58 | 48.95 | 50.2 | 51.33 | 50.56 | 50.75 | 47.6 | 49.71 |