| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| STOP (test) | WER5.7 | 18 | 3d ago | ||
| CommonVoice (CV) (test) | WER5.8 | 18 | 3d ago | ||
| CORAAL (test) | WER10.7 | 8 | 3d ago | ||
| LRS2 (test) | WER2.6 | 8 | 3d ago | ||
| SwitchBoard (test) | Word Error Rate (WER)0.042 | 8 | 3d ago | ||
| CV-accent (test) | WER0.079 | 8 | 3d ago | ||
| Tedlium-3 (test) | WER0.7 | 8 | 3d ago | ||
| CHIME-4 (test) | WER2.8 | 8 | 3d ago | ||
| ATIS (test) | WER1.1 | 8 | 3d ago | ||
| ASR Error Correction Evaluation Set (test) | Our full model | WER16.07 | 6 | 3d ago | |
| Internal Dataset (dev) | AR model | WER10.31 | 6 | 3d ago | |
| Internal Dataset (test) | AR model | WER10.22 | 6 | 3d ago | |
| AISHELL-1 (dev) | AR model | WER3.8 | 6 | 3d ago | |
| AISHELL-1 (test) | AR model | WER4.08 | 6 | 3d ago | |
| Common Voice (Persian) SNR = 10 dB | ELN-conditioned model | WER28.02 | 4 | 3d ago | |
| Common Voice (Persian) SNR = 5 dB | ELN-conditioned model | WER32.34 | 4 | 3d ago | |
| Common Voice (Persian) Mixed Noise | ELN-conditioned model | WER24.84 | 4 | 3d ago | |
| Common Voice Persian (Clean) | WER24.06 | 4 | 3d ago | ||
| MedMCQA (test) | MedSpeak | WER28.1 | 3 | 3d ago | |
| MedQA (test) | MedSpeak | WER43.5 | 3 | 3d ago | |
| MMLU (test) | Clinical Accuracy76.7 | 3 | 3d ago | ||
| VB-DEMAND English (test) | ELN-conditioned model | Baseline WER7.93 | 2 | 3d ago |
