Abstract
A cascaded Sign Language Translation system first maps sign videos to gloss annotations and then translates glosses into a spoken languages. This work focuses on the second-stage gloss translation component, which is challenging due to the scarcity of publicly available parallel data. We approach gloss translation as a low-resource machine translation task and investigate two popular methods for improving translation quality: hyperparameter search and backtranslation. We discuss the potentials and pitfalls of these methods based on experiments on the RWTH-PHOENIX-Weather 2014T dataset.- Anthology ID:
- 2021.mtsummit-at4ssl.7
- Volume:
- Proceedings of the 1st International Workshop on Automatic Translation for Signed and Spoken Languages (AT4SSL)
- Month:
- August
- Year:
- 2021
- Address:
- Virtual
- Venue:
- MTSummit
- SIG:
- Publisher:
- Association for Machine Translation in the Americas
- Note:
- Pages:
- 60–70
- Language:
- URL:
- https://aclanthology.org/2021.mtsummit-at4ssl.7
- DOI:
- Cite (ACL):
- Xuan Zhang and Kevin Duh. 2021. Approaching Sign Language Gloss Translation as a Low-Resource Machine Translation Task. In Proceedings of the 1st International Workshop on Automatic Translation for Signed and Spoken Languages (AT4SSL), pages 60–70, Virtual. Association for Machine Translation in the Americas.
- Cite (Informal):
- Approaching Sign Language Gloss Translation as a Low-Resource Machine Translation Task (Zhang & Duh, MTSummit 2021)
- PDF:
- https://preview.aclanthology.org/auto-file-uploads/2021.mtsummit-at4ssl.7.pdf
- Data
- PHOENIX14T