Achievements of the PRINCIPLE Project: Promoting MT for Croatian, Icelandic, Irish and Norwegian

Achievements of the PRINCIPLE Project: Promoting MT for Croatian, Icelandic, Irish and Norwegian Petra Bago,1 Sheila Castilho,2 Jane Dunne,2 Federico Gaspari,2 Andre Kåsen,3 Gauti Kristmannsson,4 Jon Arild Olsen,3 Natalia Resende,2 Níels Rúnar Gíslason,4 Dana D. Sheridan,5 Páraic Sheridan,5 John Tinsley,5 Andy Way2 1 Faculty of Humanities and Social Sciences, University of Zagreb, 10000 Zagreb, Croatia 2 ADAPT Centre, School of Computing, Dublin City University, Dublin 9, Ireland 3 National Library of Norway, Henrik Ibsens gate 110, 0203 Oslo, Norway 4 University of Iceland, Saemundargata 2, 102 Reykjavik, Iceland 5 Iconic Translation Machines, Invent Building, Dublin City University, Dublin 9, Ireland Abstract This paper provides an overview of the main achievements of the completed PRINCIPLE project, a 2-year action funded by the European Commission under the Connecting Europe Facility programme. PRINCIPLE focused on collecting high-quality language resources for Croatian, Icelandic, Irish and Norwegian, which are severely low- resource languages, especially for building effective machine translation (MT) systems. We report the achievements of the project, primarily in terms of the large amounts of data collected for all four low-resource languages, and of promoting the uptake of neural MT for these languages. 1. Background PRINCIPLE was a 2-year EU-funded project that ran between 2019 and 2021 to identify, collect and curate high-quality language resources (LRs) for the under-resourced languages of Croatian, Irish, Norwegian and Icelandic. The action was coordinated by the ADAPT Centre at Dublin City University (DCU), and involved the University of Iceland, the Faculty of Humanities and Social Sciences of the University of Zagreb, the National Library of Norway, and Machine Translation (MT) provider Iconic Translation Machines Ltd (now Language Weaver). The focus of the project was on providing data to improve the two Digital Service Infrastructures (DSIs) of eJustice and © 2022 The authors. This article is licensed under a Creative Commons 3.0 licence, no derivative works, attribution, CC- BY-ND. eProcurement, due to their strategic importance across the EU, in individual European Member States and in the associated countries of Iceland and Norway. Way and Gaspari (2019) introduced the PRINCIPLE project at its start, giving a high-level overview of its main objectives, along with the planned activities and the overall approach to data collection and validation. They also explained its position within the wider eco-system of related, recently finished Connecting Europe Facility (CEF) projects such as iADAATPA (Castilho et al., 2019), ELRI2 and Paracrawl.3 This paper summarises the results from PRINCIPLE, focusing on its achievements, especially in terms of engagement with stakeholders and MT users, which promoted the continued collection of LRs with a view to improving and extending MT use. 2. Achievements State-of-the-art domain-adapted neural MT (NMT) engines were built by the project partner Iconic for a number of early adopters (EAs) in all four countries. These public sector EAs included the Ministry of Foreign and European Affairs of the Republic of Croatia, the Icelandic Meteorological Office, the Icelandic Standards organisation, the Ministry of Foreign Affairs of Iceland, the Department of Justice in Ireland, Foras na Gaeilge, Rannóg an …

Content truncated to 500 words