Audio temporal grounding on Clotho-Moment
93.6R@0.3SpotSound-Q
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| SpotSound-QModel Category=Open-Source Models, Integration Backbone=Qwen2 Audio2026.04 | 93.6 | 91.2 | 85.4 | |
| SpotSound-AModel Category=Open-Source Models, Integration Backbone=Audio Flamingo 32026.04 | 93.4 | 91 | 85.6 | |
| AM-DETRModel Category=Non-LLM Models2026.04 | 89.8 | 88 | 80.9 | |
| Gemini-2.5-FlashModel Category=Proprietary Models2026.04 | 45.7 | 33.7 | 36.9 | |
| Gemini-2.5-ProModel Category=Proprietary Models2026.04 | 40.4 | 32.5 | 32.5 | |
| TimeAudioModel Category=Open-Source Models2026.04 | 39.5 | 24.9 | 28.6 | |
| Audio Flamingo 3Model Category=Open-Source Models2026.04 | 32.9 | 21.8 | 22.6 | |
| WTATGModel Category=Non-LLM Models2026.04 | 12.1 | 6.3 | 9.1 | |
| Qwen2-AudioModel Category=Open-Source Models2026.04 | 6.1 | 1.3 | 5.7 | |
| Kimi-AudioModel Category=Open-Source Models2026.04 | 0.7 | 0.1 | 0.9 |