The Swedish Sign Language (STS) Corpus mainly contains segmentations on the lexical level (i.e. signs), which makes it difficult to extract information at clause- or utterance-like levels. In this paper, I evaluate three different methods of segmenting the data into larger units: prosodic, syntactic and translation-based utterance units. The results show that none of the utterance units have particularly high accuracy in their alignment with the others, illustrating the challenges facing researchers who are looking to extract meaningful units above the lexical level. In a second step, I extract articulation information from the corpus videos using computer vision methods, but find no clear alignment of articulatory features of the hands and head with the boundaries of the utterance units.
@inproceedings{borstell:24003:sign-lang:lrec,
author = {B{\"o}rstell, Carl},
title = {Evaluating the Alignment of Utterances in the {Swedish} {Sign} {Language} Corpus},
pages = {22--31},
editor = {Efthimiou, Eleni and Fotinea, Stavroula-Evita and Hanke, Thomas and Hochgesang, Julie A. and Mesch, Johanna and Schulder, Marc},
booktitle = {Proceedings of the {LREC-COLING} 2024 11th Workshop on the Representation and Processing of Sign Languages: Evaluation of Sign Language Resources},
maintitle = {2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation ({LREC-COLING} 2024)},
publisher = {{ELRA Language Resources Association (ELRA) and the International Committee on Computational Linguistics (ICCL)}},
address = {Torino, Italy},
day = {25},
month = may,
year = {2024},
isbn = {978-2-493814-30-2},
language = {english},
url = {https://www.sign-lang.uni-hamburg.de/lrec/pub/24003.pdf}
}
