Automatic acquisition of machine translation resources in the Abu-MaTran project
Miquel Esplà-Gomis, Nikola Ljubesic, Filip Klubicka+11 more
Prokopis Prokopidis, Vassilis Papavassiliou, Antonio Toral, Tommi Pirinen, Andy Way, Raphael Rubino, Gema Ramírez-Sánchez, Sergio Ortiz-Rojas, Víctor Sánchez-Cartagena, Jorge Ferrández-Tordera, Mikel Forcada
This paper provides an overview of the research and development activities carried out to alleviate the language resources’ bottleneck in machine translation within the Abu-MaTran project. We have dev...Read more
eloped a range of tools for the acquisition of the main resources required by the two most popular approaches to machine translation, ie statistical (corpora) and rule-based models (dictionaries and rules). All these tools have been released under open-source licenses and have been developed with the aim of being useful for industrial exploitation.
Proceedings of the 1st Workshop on Open Community-Driven Machine Translation
Miquel Esplà-Gomis, Mikel L Forcada, Taja Kuzman+5 more
Nikola Ljubešić, Rik van Noord, Gema Ramírez‐Sánchez, Jörg Tiedemann, Antonio Toral
Proceedings of the 1st Workshop on Open Community-Driven Machine Translation Page 1 Proceedings of the 1st Workshop on Open Community-Driven Machine Translation June 15 2023 Tampere, Finland Edited by...Read more
Miquel Espl`a-Gomis (Universitat d’Alacant, Spain), Mikel L. Forcada (Universitat d’Alacant, Spain), Taja Kuzman (Jozef Stefan Institute, Slovenia), Nikola Ljubešic (University of Ljubljana, Slovenia), Rik van Noord (University of Groningen, The Netherlands), Gema Ramırez-Sánchez (Prompsit Language Engineering, Spain), Jörg Tiedemann (University of Helsinki, Finland), Antonio Toral (University of Groningen, The Netherlands) Organised by Page 2 Page 3 The papers published in this proceedings are —unless indicated otherwise— covered by the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC-BY-ND 4.0). You may copy, distribute, and transmit the work, …
MutNMT, an open-source NMT tool for educational purposes
We present MutNMT, 1 an open-source web application for educational purposes to introduce non-experts to NMT. The tool, developed within the MultiTraiNMT project2 along with other training materials (...Read more
a book3 and activities4), gathers the feedback of academic and industrial project partners and also external collaborators. MutNMT is based on open-source code from JoeyNMT (Kreutzer et al., 2019), 5 an open-source minimalist neural machine translation toolkit also for educational purposes. It uses its Transformer architecture to train NMT models. A new feature to extract the n-best list of translation candidates was contributed to JoeyNMT from MutNMT. MutNMT provides a user-friendly interface to manage the full process of building an NMT system, provided that training data is available, through different sections: 1) data set uploader and library for own and shared data sets, 2) engines library providing access …
Same-language machine translation for local flavours/flavors
How does AltLang work? The basics… 1/3 ● automatically and quickly replaces differences among two variants of the same language→ nice for dynamic content ● performs only controlled changes→ no (or low...Read more
) risks● highly customisable→ can adapt to DNT, lexical choices, etc. ● easily accessible→ full integration (JSON API), out-of-the-box testing (web-based demo), professional use (CAT tools and CMS)
Abu-MaTran: Automatic building of Machine Translation.
Antonio Toral, Sergio ORTIZ_ROJAS, Mikel FORCADA+2 more
Nikola Ljubesic, Prokopis Prokopidis
We present the current status of Abu-MaTran (http://www. abumatran. eu), a 4-year project (January 2013-December 2016) on rapid development of machine translation for underresourced languages. It is f...Read more
unded under Marie Curie's Industry-Academia Partnerships and Pathways 2012 programme. This is a consortium-based project with 5 partners (4 academic and 1 industrial).
AltLang: an automatic converter between varieties of English, Spanish, French and Portuguese.
AltLang is a rule-based automatic converter for language varieties. It deals with differences in spelling, lexicon and local grammar along with numeric, style and punctuation conventions. It is availa...Read more
ble for varieties of English, Spanish, French and Portuguese. AltLang is based on the GNU GPL-based free/open-source technologies of the Apertium platform, and is offered as a service by Prompsit. It can be tested through an online application at www. altlang. net.
Proceedings of the 18th Annual Conference of the European Association for Machine Translation
Ilknur Durgar El-Kahlout, Mehmed Özkan, Felipe Sánchez‐Martínez+3 more
Gema Ramírez‐Sánchez, Fred Hollowood, Andy Way
It has been a huge honour for me to serve as president of the European Association for Machine Translation (EAMT) over the past six years. As I step down from office, I am delighted that the last EAMT...Read more
annual conference under my presidency is being held in the beautiful location of Antalya, Turkey. This continues the policy I started in 2009 of bringing EAMT to new regions of Europe. This began with our first visit to the Iberian peninsula (Barcelona, 2009), our first conference in France in 2010 (St. Raphaël), followed by the Benelux region in 2011 (Leuven), continuing in 2012 with our first conference hosted in Italy (Trento), followed by the 2014 meeting in Croatia (Dubrovnik). Quite coincidentally, you may have noticed that we have journeyed step-by-step from West to East, almost 4,000 km in fact! Just to give you a little hint, next year's conference starts to reverse this trend, but only just! Of course, you'll have to wait until the closing session of this year's meeting before you find out the actual 2016 conference location!
Establishing a Linguistic Olympiad in Spain, Year 1
Antonio Toral, Guillermo Latour, Stanislav Gurevich+2 more
Mikel Forcada, Gema Ramírez-Sánchez
aims to establish a linguistic Olympiad in Spain. We introduce the Linguistic Olympiads, our rationale and objectives for setting up OLE as well as our implementation plan for. We foresee our work to ...Read more
be useful for other countries looking to start a Linguistic Olympiad.
Quality Estimation for Synthetic Parallel Data Generation.
Antonio Toral, Guillermo Latour, Stanislav Gurevich+2 more
Mikel Forcada, Gema Ramírez-Sánchez
This paper presents a novel approach for parallel data generation using machine translation and quality estimation. Our study focuses on pivot-based machine translation from English to Croatian throug...Read more
h Slovene. We generate an English–Croatian version of the Europarl parallel corpus based on the English–Slovene Europarl corpus and the Apertium rule-based translation system for Slovene–Croatian. These experiments are to be considered as a first step towards the generation of reliable synthetic parallel data for under-resourced languages. We first collect small amounts of aligned parallel data for the Slovene–Croatian language pair in order to build a quality estimation system for sentence-level Translation Edit Rate (TER) estimation. We then infer TER scores on automatically translated Slovene to Croatian sentences and use the best translations to build an English–Croatian statistical MT system. We show significant improvement in terms of automatic metrics obtained on two test sets using our approach compared to a random selection of synthetic parallel data.
Incorporating Subject Areas into the Apertium Machine Translation System
Jordi Duran, Lluís Villarejo, Mireia Farrús+2 more
Sergio Ortiz, Gema Ramírez
The Universitat Oberta de Catalunya (Open University of Catalonia, UOC), is a public university based in Barcelona. The UOC is characterised by three main factors: (a) it is a virtual university based...Read more
in an e-Learning model, (b) it is based in a strongly Spanish-Catalan bilingual region, and (c) students come from around the world, so that linguistic and cultural diversity is a crucial factor. Within this context, it becomes essential to meet the UOC’s linguistic needs taking into account its particular characteristics. One of the tools created to this end is the adaptation of Apertium, a free/open-source rule-based machine translation platform, which can be found under http://apertium.uoc.edu/, customised to the translation needs of the institution in order to offer the best possible service to their user community.