From open parallel corpora to public translation tools: The success story of OPUS

Forskningsoutput: Kapitel i bok/rapport/konferenshandlingKapitelVetenskaplig

Sammanfattning

This paper describes the success of OPUS, starting from a small side-project but leading to a full-fledged ecosystem for training and deploying open machine translation systems. We briefly present the current state of the framework focusing on the mission of increasing language coverage and translation quality in public translation models and tools that can easily be integrated in end-user applications and professional workflows. OPUS now provides the biggest hub of freely available parallel data and thousands of open translation models have been released supporting hundreds of languages in various combinations.
Originalspråkengelska
Titel på värdpublikationLIVE and LEARN : Festschrift in honor of Lars Borin
RedaktörerElena Volodina, Dana Dannélls, Aleksandrs Berdicevskis, Markus Forsberg, Shafqat Virk
Antal sidor6
UtgivningsortGöteborg
FörlagUniversity of Göteborg
Utgivningsdatumnov. 2022
Sidor133-138
ISBN (tryckt)978-91-87850-82-0
ISBN (elektroniskt)978-91-87850-83-7
StatusPublicerad - nov. 2022
MoE-publikationstypB2 Del av bok eller annan forskningsbok

Publikationsserier

NamnResearch Reports from the Department of Swedish, Multilingualism, Language Technology
FörlagUniversity of Göteborg
NummerGU-ISS-2022-03
ISSN (elektroniskt)1401-5919

Vetenskapsgrenar

  • 6121 Språkvetenskaper
  • 113 Data- och informationsvetenskap

Citera det här