Embracing LangOps: Mandatory in the Generative AI Era
Jochen Hummel has a software development background and has grown his first company, TRADOS, to the world leader in translation memory and terminology software. In 2006 he founded Metaversum, the inventor of Twinity, the online virtual world based on real cities, and was its CEO until 2010.
The language industry experienced its ChatGPT moment seven years ago with the advent of Neural Machine Translation (NMT) claiming human parity. These unique experiences are now proving invaluable, positioning language professionals for a key role in generative AI. Language Operations is an innovative methodology empowering businesses to adeptly manage their multilingual data. The traditional task of “doing words” is nearing its end, superseded by establishing robust data pipelines to bolster AI initiatives – a process that is not just essential and value-creating but also inherently engaging.
GenAI: the next step in the evolution of translation?
Constantin Orasan is a Professor of Language and Translation Technologies at the Centre for Translation Studies, University of Surrey, and a Fellow of the Surrey Institute for People-Centred Artificial Intelligence. Before starting the job at the University of Surrey, he was a Reader in Computational Linguistics at the University of Wolverhampton, UK, and the deputy head of the Research Group in Computational Linguistics at the same university. He has over 20 years of experience in the fields of Natural Language Processing (NLP), Translation Technologies, Artificial Intelligence, and Machine Learning for language processing. He is very interested in large language models, like ChatGPT and BARD, and how they impact translation and the translation industry. This will be the topic of his talk at HiT-IT 2023.
Towards Responsible Machine Translation with Quality Estimation
Dr. Tharindu Ranasinghe is a lecturer in Computer Sciences at the School of Engineering and Applied Science, Aston University, UK. His research focuses on various aspects of machine learning-driven approaches to natural language processing, with a particular interest in multilingual models and explainable machine learning. His work has significant applications including translation quality estimation, social media data mining, offensive language identification, information extraction, and digital humanities.
Live speech-to-text and machine translation tool for 24 languages – increasing accessibility of parliamentary debates.
Marcin Feder has a Ph.D. in linguistics (English/Computer Assisted Translation). He worked for Adam Mickiewicz University in Poznań, Poland, as an assistant professor. Then in 2003, he joined the EP as an in-house interpreter in the Polish booth. Later he became the Head of the PL Unit in DG LINC, and subsequently Head of the Interpreter Support and Training Unit and acting Head of the Multilingualism and Succession Planning Unit. Since 2019, Marcin is the Head of the Speech-to-text Unit in DG TRAD, which is developing a live speech-to-text and machine translation tool that can automatically transcribe and translate parliamentary multilingual debates in real-time. The ultimate goal of this project, which Marcin will talk to us about, is to increase accessibility to and provide an automatic transcription and translation service for parliamentary debates covering the 24 official languages used by the European Parliament.
The Importance of Data Anonymization to Build Ethical AI
Manuel Herranz, Pangeanic CEO
Manuel is Pangeanic CEO and MIT Sloan School of Management graduate. He is a language technology veteran with 25 years industry experience. His concept of DIY Statistical MT underpins current approaches for MT customisation by Bing Translator, Google AutoML, ModernMT and Amazon’s self-training mechanisms. He has written or co-written nearly 20 academic papers – several being the result of the numerous EU projects he’s led or been involved in as a partner.
Data provenance and the use and inclusion of personal data have not been historically a consideration in the building of machine translation systems and, lately, Large Language Models (LLMs).
Manuel’s keynote will touch upon concepts such as Ethical AI and the need to build ethics all along the data generation and data acquisition pipelines when building AI products and services, GDPR, compliance, EU anonymisation projects, the data-for-AI market, etc.
Smart concordance search on EU terminology integrated into Trados Studio – the Juremy Web Plugin
Tímea Palotai-Torzsás, Co-Founder of Juremy.com
Tímea is a lawyer specializing in intellectual property matters, and a legal translator (EN-HU, FR-HU) with 8 years of experience as a freelance lawyer-linguist for the Court of Justice of the European Union. She has been involved in legal translation projects since the beginning of her professional activity as a lawyer. She is the co-founder of the online application Juremy.com, which was launched in 2019 to support linguists’ EU terminology research workflow. She is passionate about legal and terminology research, legal language and translation productivity. She also holds a bachelor degree in international communication.
Juremy.com is an online concordance search tool which provides fast and customizable bilingual phrase search on the pre-translated segments from the EUR-Lex corpus and the IATE termbase. By offering an innovative time-saving solution in terminology research, the tool helps enhance quality by bringing bilingual contextual search results in less than a second in all 24 official EU-languages. Juremy offers EU-translation specific functionalities such as sorting and filtering on document metadata, ranking hits based on topic preferences, displaying results by smart search categories including fuzzy match hits. A web-based CAT-tool integration is also available which enables users to access the Juremy browser interface within Trados Studio and to use Juremy interactively when translating or reviewing a document.
Juremy was created by a lawyer-linguist with several years of experience as a freelancer for the Court of Justice of the European Union, and a software-engineer passionate about building user-centric features. With a view to design a terminology search tool for the specific workflow of translators and interpreters working with legal and EU-related texts, the founders launched Juremy.com in 2019, and they are continuously working on adding new functionalities to the tool based on users’ needs. Juremy is mostly used by freelance and institutional translators, terminologists, interpreters and LSPs working with EU-related projects.
sTMS Cloud – a Boutique Translation Project Management System by Mitra Translations
DIlyana Ilieva, CEO @Mitra Translations
Mina Ilieva, Managing Director @Mitra Translations
Dilyana’s input is invaluable when handling both client relations and managing personnel. A quick turnarounder with strong skills in technical and software solutions, she can prepare you a great build up plan for every ongoing or future project.
Mina has a professional practice in the translation industry for more than 25 years and is operating as a Managing Director of MiTRa since 2004. She has full technology expertise leading to both business growth and stability. Her sharp mind results in a boosted company profile.
Cloud-based Translation Project Management System, called sTMS, developed with the financial support of Operational Programme “Innovation and Competitiveness” 2014-2020 (OPIC) focusing to enhance the operational activities of LSPs and MLPs. The idea behind was to concentrate mainly on the management processes, and not to integrate CAT or MT tools, because we believe that the more functional such systems become, the harder to technically support and easy to operate they become. The key features sTMS provides are developed as a result of the broad experience of our Project Managers, the increased requirements of our customers, the digital capabilities of our vendors and as last to meet the constantly changing environment of the translation industry.