Speech-to-Text Translation on FLEURS zero-shot evaluation languages (test)
20.84Average BLEUPOTSA (Our Model)
Evaluation Results
| Method | Links | |
|---|---|---|
| POTSA (Our Model)evaluation_mode=zero-shot, Bias Compensation=true, OT Alignment=true2025.11 | 20.84 | |
| POTSA (w/o Bias Comp.)evaluation_mode=zero-shot, Bias Compensation=false2025.11 | 19.34 | |
| POTSA (w/o OT Align.)evaluation_mode=zero-shot, OT Alignment=false2025.11 | 18.51 | |
| POTSA (En-Anchor Fixed)evaluation_mode=zero-shot, Anchor Strategy=Fixed2025.11 | 18.37 | |
| WhisperV3+Qwen2.5Training Approach=end-to-end, evaluation_mode=zero-shot2025.11 | 17.91 | |
| Qwen2.5-Omnievaluation_mode=zero-shot2025.11 | 16.91 | |
| Qwen2-Audioevaluation_mode=zero-shot2025.11 | 16.67 | |
| POTSA (En-Anchor Train.)evaluation_mode=zero-shot, Anchor Strategy=Trainable2025.11 | 11.53 |