Speech Recognition with Timestamps on TIMIT
34.3AAS ScoreCrisperWhisper
Evaluation results
| Method | Links | ||
|---|---|---|---|
| CrisperWhisper2026.04 | 34.3 | — | |
| Qwen-Audio2026.04 | 22.25 | 6 | |
| Mixed Training + Length AugLength Augmentation=true, Evaluation Setting=zero-shot2026.04 | 20.48 | 0 | |
| Mixed TrainingEvaluation Setting=zero-shot2026.04 | 20.23 | 0 | |
| Mixed Training + Length Aug + Timestamp RegLength Augmentation=true, Timestamp Regularization=true, Evaluation Setting=zero-shot2026.04 | 20.22 | 0 | |
| Mixed Training + Length Aug + Reduced TFLength Augmentation=true, Reduced Teacher Forcing=true, Evaluation Setting=zero-shot2026.04 | 19.89 | 0 |