Context-Aware Video Speech Recognition (CAVSR) on VASR (test)
11.02CERVASR
Evaluation Results
| Method | Links | |
|---|---|---|
| VASRParameters=7B, Input Modality=A/V2026.03 | 11.02 | |
| Gemini2.5ProParameters=-, Input Modality=A/V2026.03 | 11.81 | |
| Qwen3Omni-Instruct-30B-A3BParameters=30B, Input Modality=A/V2026.03 | 11.97 | |
| Qwen2.5Omni-7BParameters=7B, Input Modality=A/V2026.03 | 12.21 | |
| Qwen3Omni-Thinking-30B-A3BParameters=30B, Input Modality=A/V2026.03 | 12.39 | |
| Qwen2.5Omni-3BParameters=3B, Input Modality=A/V2026.03 | 13.06 | |
| Doubao ASRParameters=-, Input Modality=A2026.03 | 17.17 | |
| Intern-S1Parameters=9B, Input Modality=A/V2026.03 | 17.6 | |
| Qwen2.5Omni-7BParameters=7B, Input Modality=A2026.03 | 19.25 | |
| MiniCPM-o2.6Parameters=8B, Input Modality=A/V2026.03 | 19.49 |