Audio-to-Text Generation on one-to-one evaluation benchmarks
45.08CLAP ScoreOmniFlow
Evaluation Results
| Method | Links | |
|---|---|---|
| OmniFlowCategory=Generalists2025.12 | 45.08 | |
| FlowBindCategory=Generalists2025.12 | 36.7 | |
| UnifiedIO2-LCategory=Generalists2025.12 | 18.68 | |
| CoDiCategory=Generalists2025.12 | 17.94 | |
| Qwen2-AudioCategory=Specialists2025.12 | 17.09 |