Automatic Speech Recognition on Prospective Multi-Center Endoscopy Dataset (C4)
10.97CEREndoASR-noise
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| EndoASR-noiseDescription=Domain-adapted model after stage-1 and stage-2 fine-tuning2026.04 | 10.97 | 89.26 | 94.31 | 88.07 | |
| EndoASRDescription=Domain-adapted model after stage-1 and stage-2 fine-tuning2026.04 | 11.2 | 89.01 | 94.44 | 85.83 | |
| ParaformerDescription=Baseline model2026.04 | 12.34 | 87.97 | 93.69 | 66.44 |