Multilingual Gloss-free Sign Language Translation: Towards Building a Sign Language Foundation Model
This work addresses the communication gap between sign and spoken communities by enabling translation across multiple sign languages, though it is incremental as it builds on existing SLT methods.
The paper tackles multilingual sign language translation (MLSLT) by proposing a gloss-free model with dual CTC objectives, achieving competitive performance on three benchmarks including multilingual SP-10, PHOENIX14T, and CSL-Daily.
Sign Language Translation (SLT) aims to convert sign language (SL) videos into spoken language text, thereby bridging the communication gap between the sign and the spoken community. While most existing works focus on translating a single sign language into a single spoken language (one-to-one SLT), leveraging multilingual resources could mitigate low-resource issues and enhance accessibility. However, multilingual SLT (MLSLT) remains unexplored due to language conflicts and alignment difficulties across SLs and spoken languages. To address these challenges, we propose a multilingual gloss-free model with dual CTC objectives for token-level SL identification and spoken text generation. Our model supports 10 SLs and handles one-to-one, many-to-one, and many-to-many SLT tasks, achieving competitive performance compared to state-of-the-art methods on three widely adopted benchmarks: multilingual SP-10, PHOENIX14T, and CSL-Daily.