TTIC’s WMT-SLT 22 Sign Language Translation System

Bowen Shi, Diane Brentari, Gregory Shakhnarovich, Karen Livescu


Abstract
We describe TTIC’s model submission to WMT-SLT 2022 task on sign language translation (Swiss-German Sign Language (DSGS) - German). Our model consists of an I3D backbone for image encoding and a Transformerbased encoder-decoder model for sequence modeling. The I3D is pre-trained with isolated sign recognition using the WLASL dataset. The model is based on RGB images alone and does not rely on the pre-extracted human pose. We explore a few different strategies for model training in this paper. Our system achieves 0.3 BLEU score and 0.195 Chrf score on the official test set.
Anthology ID:
2022.wmt-1.96
Volume:
Proceedings of the Seventh Conference on Machine Translation (WMT)
Month:
December
Year:
2022
Address:
Abu Dhabi, United Arab Emirates (Hybrid)
Venue:
WMT
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
989–993
Language:
URL:
https://aclanthology.org/2022.wmt-1.96
DOI:
Bibkey:
Cite (ACL):
Bowen Shi, Diane Brentari, Gregory Shakhnarovich, and Karen Livescu. 2022. TTIC’s WMT-SLT 22 Sign Language Translation System. In Proceedings of the Seventh Conference on Machine Translation (WMT), pages 989–993, Abu Dhabi, United Arab Emirates (Hybrid). Association for Computational Linguistics.
Cite (Informal):
TTIC’s WMT-SLT 22 Sign Language Translation System (Shi et al., WMT 2022)
Copy Citation:
PDF:
https://preview.aclanthology.org/emnlp-22-attachments/2022.wmt-1.96.pdf