0

SignBank+: Preparing a Multilingual Sign Language Dataset for Machine Translation Using Large Language Models

The SignBank+ dataset, a cleaned version of SignBank, supports effective machine translation between text and SignWriting using traditional text-to-text translation methods, surpassing the original dataset's performance.

Year
2023
Venue
arXiv 2023
Authors
2
Hosting
Abstract onlyARXIV-DEFAULT

Cite

Notes

Only stored in your browser.

Attribution

Abstract & full text
arxiv.org/abs/2309.11566v2ARXIV-DEFAULT
TL;DR
Semantic Scholar
Attribution policy →

Abstract

We introduce SignBank+, a clean version of the SignBank dataset, optimized for machine translation between spoken language text and SignWriting, a phonetic sign language writing system. In addition to previous work that employs complex factorization techniques to enable translation between text and SignWriting, we show that a traditional text-to-text translation approach performs equally effectively on the cleaned SignBank+ dataset. Our evaluation results indicate that models trained on SignBank+ surpass those on the original dataset, establishing a new benchmark for SignWriting-based sign language translation and providing an open resource for future research.

Authors

2