Products & Projects

Misinformation on social media is a concern for content creators, consumers and regulators alike. Transitude looks at misinformation generated by machine translation (MT) through distortion of the intention and sentiment of text. It is the first study of MT’s impact on the formation of users’ views of society through refugees in Ireland. It extends current MT evaluation methods with a new quality evaluation framework, producing the first dataset annotated for information distortion. It provides insights into the risks of relying on MT, with recommendations for users, developers, and policymakers.

The LiLowLa (“Lightweight neural translation technologies for low-resource languages”) project aims to enhance machine translation (MT) and translation memory (TM) technologies, particularly for low-resource language pairs, where adequate linguistic resources are scarce. The project started in September 2022 and will run till August 2025.

This project aims to develop a multilingual notification system for asylum reception centres in Belgium using machine translation. The system will allow staff to communicate practical messages to residents in their own language. Ethnographically inspired fieldwork is being conducted in reception centres to understand current communication practices and ensure that the technology meets user needs. The quality and suitability of machine translation will be evaluated for three MT systems supporting all target languages. Automatic and manual evaluation methods will be used to assess translation quality, and terms of use, privacy and data protection conditions will be analysed.

SmartBiC, an 18-month innovation project funded by the Spanish Government, aims at improving the full process of collecting, filtering and selecting in-domain parallel content to be used for machine translation and language model tuning purposes in industrial settings. Based on state-of-the-art technology in the free/open-source parallel web corpora harvester Bitextor, SmartBic develops a web-based application around it including novel components such as a language- and domain-focused crawler and a domain-specific corpora selector. SmartBic also addresses specific industrial use cases for individual components of the Bitextor pipeline, such as parallel data cleaning. Relevant improvements to the current Bitextor pipeline will be publicly released.

This EAMT-funded eye-tracking study investigates the impact of Machine Translation Post-Editing and Automatic Speech Recognition on English-Romanian translations of patient-facing medical texts. This paper provides an overview of the study objectives, setup and preliminary results.

This paper describes MAKE-NMTViz, a project designed to help translators visualize neural machine translation outputs using explainable artificial intelligence visualization tools initially developed for computer vision.

In this paper, we present the MULTILINGTOOL project, led by the Elhuyar Foundation and funded by the European Commission under the CREA-MEDIA2022-INNOVBUSMOD call. The aim of the project is to develop an advanced platform for automatic multilingual subtitling and dubbing. It will provide support for Spanish, English, and French, as well as the co-official languages of Spain, namely Basque, Catalan, and Galician.

The CALCULUS project, drawing on human capabilities of imagination and commonsense for natural language understanding (NLU), aims to advance machine-based NLU by integrating traditional AI concepts with contemporary machine learning techniques. It focuses on developing anticipatory event representations from both textual and visual data, connecting language structure to visual spatial organization and incorporating broad knowledge domains. Through testing these models in NLU tasks and evaluating their ability to predict untrained spatial and temporal details using real-world metrics, CALCULUS employs machine learning methods, including Bayesian techniques and neural networks, especially in data-sparse scenarios. The project’s culmination involves creating demonstrators that transform written stories into dynamic videos, showcasing the interdisciplinary expertise of the project leader in natural language processing, language and visual data analysis, information retrieval, and machine learning, all vital for the project’s achievements. In the CALCULUS project, our exploration of machine translation extends beyond the conventional text-to-text framework. We are broadening the horizons of machine translation by delving into the essence of transforming the formats of data distribution while keeping the meaning. This innovative approach involves converting information from one modality into another, transcending traditional linguistic boundaries. Our project includes novel work on translating text into images and videos, brain signals into images and videos.

The «App for post-editing neural machine translation using gamification» (GAMETRAPP) project (TED2021-129789B-I00), funded by the Spanish Ministry of Science and Innovation (2022–2024), has been in progress for a year. Thus, this paper presents its main goals and the analysis of neural machine translation and post-editing errors of research abstracts carried out. This leads to the designing of the gamified environment, which is currently under construction.

The RCnum project is funded by the Swiss National Science Foundation and aims at producing a multilingual and semantically rich online edition of the Registers of Geneva Council from 1545 to 1550. Combining multilingual NLP, history and paleography, this collaborative project will clear hurdles inherent to texts manually written in 16th century Middle French while allowing for easy access and interactive consultation of these archives.

This article presents the main functionality of the postedit.me app. Postedit.me is a software program that supports machine translation post-editing training in translator education, with special emphasis on standardized quality evaluation of post-edited texts produced by students. The app is made freely available to universities for teaching and research purposes.

Artificial intelligence (AI) is quickly becoming an exciting new technology for the translation industry in form of large language models (LLMs). AI-based functionality could be used to improve the output of neural machine translation (NMT). One main issue that impacts MT quality and reliability is incorrect terminology. This is why STAR is making AI-powered terminology control a priority for its translation products because of the significant gains to be made - greatly improving the quality of MT output, reducing post editing (PE) costs and efforts, and thereby boosting overall translation productivity.

This research project aims to develop a comprehensive methodology to help make machine translation (MT) systems more gender-inclusive for society. The goal is the creation of a detection system, a machine learning (ML) model trained on manual annotations, that can automatically analyse source data and detect and highlight words and phrases that influence the gender bias inflection in target translations.The main research outputs will be (1) a manually annotated dataset, (2) a taxonomy, and (3) a fine-tuned model.

The INCREC project aims to uncover professional translators’ creative stages to understand how technology can be best applied to the translation of literary and audio-visual texts, and to analyse the impact of these processes on readers and viewers. To better understand this process, INCREC triangulates data from eye-tracking, retrospective think-aloud inter-views, translated material, and questionnaires from professional translators and users.

We introduce SMUGRI-MT, an online neural machine translation system that covers 20 low-resource Finno-Ugric languages, along with seven high-resource languages.

plain X is a 4-in-1 solution for language adaptation. The software is an outcome of European HLT research and is by now in use as the major artificial-intelligence-powered human language pro-cessing platform at Deutsche Welle. plain X is a one-stop-shop for automated transcription, translation, subtitling and voice-over, with human correction options at all stages. We demonstrate how the platform works and show new features and developments of the platform in the framework of the SELMA project.

This paper describes the project “BridgeAI: Boosting Regulatory Implementation with Data-driven insights, Global expertise, and Ethics for AI”, a one-year science-for-policy research project funded by the Portuguese Foundation for Science and Technology (FCT). The project aims to provide decision-makers in Portugal with the best context to implement the EU Artificial Intelligence (AI) Act and bridge the gap between AI research and policy. Although not exclusively on machine translation, the project pertains to natural language processing in general and ultimately to each of us as citizens.

Research on gender bias in Machine Translation (MT) predominantly focuses on binary gender or few languages. In this project, we investigate the ability of commercial MT systems and neural models to translate using gender-fair language (GFL) from English into German. We enrich a community-created GFL dictionary, and sample multi-sentence test instances from encyclopedic text and parliamentary speeches. We translate our resources with different MT systems and open-weights models. We also plan to post-edit biased outputs with professionals and share them publicly. The outcome will constitute a new resource for automatic evaluation and modeling gender-fair EN-DE MT.

Addressing online disinformation requires analysing narratives across languages to help fact-checkers and journalists sift through large amounts of data. The ExU project focuses on developing AI-based models for multilingual disinformation analysis, addressing the tasks of rumour stance classification and claim retrieval. We describe the ExU project proposal and summarise the results of a user requirements survey regarding the design of tools to support fact-checking.

VIGILANT (Vital IntelliGence to Investigate ILlegAl DisiNformaTion) is a three-year Horizon Europe project that will equip European Law Enforcement Agencies (LEAs) with advanced disinformation detection and analysis tools to investigate and prevent criminal activities linked to disinformation. These include disinformation instigating violence towards minorities, promoting false medical cures, and increasing tensions between groups causing civil unrest and violent acts. VIGILANT’s four LEAs require support for English, Spanish, Catalan, Greek, Estonian, Romanian and Russian. Therefore, multilinguality is a major challenge and we present the current status of our tools and our plans to improve their performance.

This paper presents a dataset for evaluating the machine translation of emotion-loaded user generated content. It contains human-annotated quality evaluation data and post-edited reference translations. The dataset is available at our GitHub repository.

Among the services provided by Softcatalà, a non-profit 25-year-old grassroots organization that localizes software into Catalan and develops software to ease the generation of Catalan content, one of the most used is its machine translation (MT) service, which provides both rule-based MT and neural MT between Catalan and twelve other languages. Development occurs in a community-supported, transparent way by using free/open-source software and open language resources. This paper briefly describes the MT services at Softcatalà: the offered functionalities, the data, and the software used to provide them.

MTxGames is a doctoral research project examining three different machine translation (MT) post-editing (PE) methods in the context of translating creative texts from video games, focusing on translation speed, cognitive effort, quality, and translators’ preferences. This is a mixed-methods study, eliciting quantitative data through keylogging, eye-tracking, and error evaluation as well as qualitative data through interviews. To create realistic experimental conditions, data elicitation takes place at the workplaces of freelancing professional game translators.

SignON, a 3-year Horizon 20202 project addressing the lack of technology and services for MT between sign languages (SLs) and spoken languages (SpLs) ended in December 2023. SignON was unprecedented. Not only it addressed the wider complexity of the aforementioned problem – from research and development of recognition, translation and synthesis, through development of easy-to-use mobile applications and a cloud-based framework to do the “heavy lifting” as well as to establishing ethical, privacy and inclusivenesspolicies and operation guidelines – but also engaged with the deaf and hard of hearing communities in an effective co-creation approach where these main stakeholders drove the development in the right direction and had the final say.Currently we are witnessing advances in natural language processing for SLs, including MT. SignON was one of the largest projects that contributed to this surge with 17 partners and more than 60 consortium members, working in parallel with other international and European initiatives, such as project EASIER and others.

In the relief operations of international humanitarian organisations, non-governmental organisations (NGOs) often encounter language needs when delivering services (Tesseur 2022). This project examines the language needs of humanitarian NGOs working from Hong Kong and the solutions they adopted to overcome the language barriers when delivering international humanitarian relief to other countries.

The High Performance Language Technologies (HPLT) project is a 3-year EU-funded project that started in September 2022. It aims to deliver free, sustainable, and reusable datasets, models, and workflows at scale using high-performance computing. We describe the first results of the project. The data release includes monolingual data in 75 languages at 5.6T tokens and parallel data in 18 language pairs at 96M pairs, derived from 1.8 petabytes of web crawls. Building upon automated and transparent pipelines, the first machine translation (MT) models as well as large language models (LLMs) have been trained and released. Multiple data processing tools and pipelines have also been made public.

LT-LiDER is an Erasmus+ cooperation project with two main aims. The first is to map the landscape of technological capabilities required to work as a language and/or translation expert in the digitalised and datafied language industry. The second is to generate training outputs that will help language and translation trainers improve their skills and adopt appropriate pedagogical approaches and strategies for integrating data-driven technology into their language or translation classrooms, with a focus on digital and AI literacy.

We present how at Unbabel we have been using Large Language Models to apply a Cultural Transcreation (CT) product on customer support (CS) emails and how we have been testing the quality and potential of this product. We discuss our preliminary evaluation of the performance of different MT models in the task of translating rephrased content and the quality of the translation outputs. Furthermore, we introduce the live pilot programme and the corresponding relevant findings, showing that transcreated content is not only culturally adequate but it is also of high rephrasing and translation quality.

The AI4Culture project (2023-2025), funded by the European Commission, and involving a 12-partner consortium led by the National Technical University of Athens, develops a platform serving as an online capacity building hub for AI technologies in the cultural heritage (CH) sector, enabling multilingual access to CH data. It offers access to AI-related resources, including openly labelled datasets for model training and testing, deployable and reusable tools, and capacity building materials. The tools are aimed at optical character recognition (OCR) for printed and handwritten documents, subtitle generation and validation, machine translation (MT), and metadata enrichment via image information extraction and semantic linking. The project also customises these tools to enhance interface and component usability. We illustrate this with technology that corrects OCR output using language models and adapts it for MT.

This paper describes the project “NextGenAI: Center for Responsible AI”, a 39-month Mobilizing and Green Agenda for Business Innovation funded by the Portuguese Recovery and Resilience Plan, under the Recovery and Resilience Facility (RRF). The project aims to create a new Center for Responsible AI in Portugal, capable of delivering more than 20 AI products in crucial areas like “Life Sciences”, many of which use generative AI, particularly NLP models such as those for Machine Translation, contributing to translating into legislation the European Law included in the EU AI Act, and creating a critical mass in the development of responsible AI technologies. To accomplish this mission, the Center for Responsible AI is formed by an ecosystem of startups and research institutions driving research in a virtuous way by addressing real market needs and opportunities in Responsible AI.