Automatic Speech Recognition on DefinedAI Insurance target domain (test)
7.92WERAdapted model (audio)
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Adapted model (audio)Adaptation=Audio-text pairs2026.01 | 7.92 | 25.4 | |
| Text Denoising AdaptationAdaptation=Text-only2026.01 | 8.71 | 17.9 | |
| Ma et al.Adaptation=Text-only2026.01 | 9.68 | 8.8 | |
| Fang et al.Adaptation=Text-only2026.01 | 9.79 | 7.7 | |
| Base modelAdaptation=None2026.01 | 10.61 | — |