Access to workplace information is a significant barrier to participation in the primary labor market for people with cognitive impairments. Although Easy Language (EL) improves accessibility, producing high-quality EL texts is resource-intensive and requires specialized expertise. Recent advances in artificial intelligence offer new opportunities to support the creation and dissemination of accessible workplace communication.
The STARK-LS project (Strengthening Participation in the Primary Labor Market through AI-Generated Easy Language) explores the potential of AI-assisted EL translation in real workplaces. Within the project, workplace materials are first translated into German Easy Language using an AI-based translation system. Then, they are post-edited according to current EL standards. These materials are then used in workplace internships involving people with cognitive impairments to evaluate linguistic quality and practical applicability.
The project incorporates various evaluation perspectives, such as assessing the quality of AI-generated and post-edited EL texts, analyzing post-editing effort, conducting user-centered studies on comprehensibility and usability, and investigating organizational acceptance among employers. Additionally, STARK-LS explores innovative approaches to delivering accessible information. To this end, a prototype mobile application has been developed that captures workplace texts, generates Easy Language versions through AI-supported processing, and provides context-sensitive access to simplified information. The application serves as a platform for investigating how digital delivery formats can support access to workplace information and task-related communication.
The poster presents the project framework, the AI-supported translation workflow, the prototype application, and the ongoing research agenda. By examining the production, evaluation, and workplace application of AI-generated and post-edited Easy Language, STARK-LS contributes to the ongoing discussions surrounding AI-assisted translation, accessibility, and inclusive communication in the algorithmic era.
Athanasios Breskas holds an MA in Translation from the Faculty of Translation Studies, Linguistics and Cultural Studies at Johannes Gutenberg University Mainz in Germersheim. He is a doctoral researcher at the TRA&Co Center and a project partner in STARK-LS. His research interests include virtual reality, augmented reality, eyetracking, keylogging, artificial intelligence, machine translation, foreign language acquisition, cognition and easy language.
Silke Gutermuth works as a postdoc research assistant and lecturer at the Faculty of Translation Studies, Linguistics, and Cultural Studies (FTSK) at the Johannes Gutenberg University Mainz in Germersheim where she obtained her B.A. and M.A. in Translation Studies. Her main research interests include translation process research and accessible communication with a focus on Easy Language. Her PhD thesis explored a multi-method approach to Easy and plain language reception. She is responsible for coordination and experimental setup in the eyetracking lab at the TRA&Co Center and co-manages the Competence Center for Translation and Communication (CCTC) in Germersheim.
Silvia Hansen-Schirra is Professor for English Linguistics and Translation Studies, the Director of the Translation & Cognition (TRA&Co) Center and Dean of the Faculty for Translation Studies at Johannes Gutenberg University Mainz in Germersheim. She is the co-editor of the book series“Translation and Multilingual Natural Language Processing” and “Easy – Plain – Accessible”. Her research interests include machine translation, accessible communication and translation process research.
Fabian Merkel is a PhD student and research associate at the Faculty of Translation Studies, Linguistics, and Cultural Studies (FTSK) at the Johannes Gutenberg University Mainz, where he obtained his B.A and M.A. in Translation Studies. His main research interests are Easy Language and the use of AI in Translation.
