Ugarit

From The Digital Classicist Wiki
Jump to navigation Jump to search

Available

Author

  • Tariq Yousef

Description

Ugarit is an online environment to perform manual translation alignment of texts. It provides an editor where users can import their corpora and align them manually, and publish them immediately in the homepage. The editor enables alignment of two or three texts, and it supports most ancient and Classical languages. Users can perform four levels of alignment: word-to-word (1-1), word-to-phrase (1-N), phrase-to-word (N-1) and phrase-to-phrase (N-N).

The Ugarit homepage provides an interactive reading environment where users can read available aligned texts, click on aligned words to trigger a dynamic lexicon, and search for already aligned words. The reader also provides an automatic transliterator for non-Latin alphabets, basic statistics on published alignments, and two export options (XML or CSV).

The Ugarit database provides training data, in the form of translation pairs, for the development of automated translation alignment models targeted at ancient languages. Most recently, Yousef et al. (2022) have published a transformer-based model that uses Ugarit alignments for fine-tuning in Ancient Greek.

See also: Translation Alignment.

References

Alignment Guidelines and Gold Standards

Presentations

Reviews

  • Sisi Xie. 2023. "Textual Alignment of Res Gestae: Translation in Historical Languages. Commenting on the Ugarit alignment tool." Stoa Review 2023-01-26. Available: https://blog.stoa.org/archives/4182