Room 3, Building R | Media for all 10

10.30 - 12.00: Session 11 - Training

Promoting accessibility and inclusion with subtitling for the deaf and hard of hearing (SDH) as an AVT tool in the EFL classroom

Anca Daniela Frumuselu, Antonio Jesús Tinedo-Rodríguez

The field of didactic Audiovisual Translation (AVT) has grown exponentially in the last twenty years due to shared efforts of researchers and practitioners who have looked at how the active task of subtitling, captioning, and revoicing among foreign language learners has impacted their foreign language and communication skills (Lertola, 2019; Talavan, 2020). Promoting social awareness, accessibility, and inclusion among foreign language learners should be of paramount importance for both teachers and practitioners given the current social issues worldwide. Therefore, audiovisual products initially aimed at people with visual impairment and hard of hearing can be adapted for any type of learner in order to make our society fairer and more inclusive. This research paper draws back on the implementation of a didactic sequence promoted through the TRADILEX project, funded by the Spanish Ministry of Science, whose main goal is to explore the pedagogical potential of the five main modalities of AVT. The main objective of this research study is to foster awareness of media accessibility and inclusion through the application of SDH as a didactic AVT resource to explore its pedagogical potential in the EFL classroom. Several instruments have been developed within the context of the TRADILEX Project: a pre-test – The Initial Test of Integrated Skills (ITIS), a post-test – The Final Test of Integrated Skills (FITIS), which have been designed by (Couto-Cantero, Sabaté-Carrové, & Gómez Pérez, 2021), a didactic sequence (6 lesson plans on SDH for B2 for the piloting phase course and 3 SDH lesson plans as part of the entire B2 course of the AVT didactic sequence), an initial questionnaire and a final questionnaire. First, students were asked to fill in the initial questionnaire and sit the ITIS test. Then, they were asked to carry out the didactic sequence, containing 6 (in the piloting phase) and 3 (in the regular data collection process) lesson plans around various topics that integrate the use of SDH in an inclusive way. The set of lesson plans (LP, hereinafter) consisted of one-hour sessions divided into four parts: warm-up, video viewing, AVT task, and post-AVT task. The main objective of the first two parts of the LP was to make students familiar with the fragment of the video they were asked to subtitle. The AVT task consisted of subtitling the video by making use of SDH and, eventually, they were provided with tasks on their newly acquired knowledge in the post-AVT activity. After the intervention, students were asked to sit the FITIS test and complete the final questionnaire. The preliminary results show that students tend to develop their EFL production skills (writing and speaking) after being exposed to SDH activities over 6 weeks and that they gain awareness of the importance of media accessibility and inclusion. Further details about the SDH evaluation rubrics and how learners’ production tasks were assessed and how they evolved over time will be presented at the conference.

Anca Daniela Frumuselu is currently working as a Lecturer of English at the University of Lleida, Spain. She obtained her double international doctoral degree in 2016 from Universitat Rovira I Virgili (Spain) and from University of Antwerp (Belgium). Her research interests include the study of English foreign language skills through audiovisual translation, subtitling and subtitling for the deaf and hard of hearing, the use of new technologies in the EFL classroom, e-learning, m-learning, gamification and virtual exchange. She has attended several international conferences, workshops and teaching training courses in the last years and she has published articles in high ranked journals related to her fields of interest. She has also participated as a research member in various European Erasmus + KA2 and Spanish funded projects. Most recently, she has participated in the I+D+i project, TRADILEX (Audiovisual Translation as a Didactic Resource in Foreign Language Education), in collaboration with other international institutions and has been investigating accessibility modalities, such as subtitling for the deaf and hard of hearing in the English foreign language classroom.

Antonio Jesús Tinedo-Rodríguez is a scholar who holds a degree in Primary Education from the University of Córdoba and a degree in English Studies from UNED. He also has a master's degree in ICT in Language Teaching and Treatment from UNED, and is currently pursuing his PhD in Didactic Audiovisual Translation. During his initial training, Antonio was awarded the University Department Collaboration Grant (MEFP), which provided him with the opportunity to begin his research career in the Department of Education at the University of Cordoba. He has also served as a research technician in the RDi TRADILEX project and was awarded a UNED-Santander predoctoral contract (2022). In addition, he received a grant from UNED-Santander for a research stay at the Università del Piemonte Orientale (UniUPO) in Italy, where he delved deeper into Didactic Audiovisual Translation and Specialised Audiovisual Texts. In 2023, he was awarded a prestigious FPI predoctoral (EIDUNED) contract for his doctoral studies at the Department of Foreign Philologies at the UNED. His research focuses on several areas, including Didactic Audiovisual Translation (DAT), language learning, bilingual and intercultural education (BIE), and discursive analysis from a gender perspective. He is particularly interested in exploring the didactic possibilities of texts written by Anglo-American female authors.

Co-creating accessible videos: the mediaverse-CROMA experience

Anna Matamala, Víctor García, Mariona González-Sordé

Mediaverse is a European project that aims to develop an innovative ecosystem where creators can share their media content while keeping control of intellectual property rights and monetization. As part of this project, different pilots are taking place in order to show how a cocreation process can be used to different ends: to produce artistic content, to produce journalistic content and to generate 360º interactive content for social inclusion. Within the context of social inclusion, Mediaverse has partnered with the CROMA project at Fundació Autònoma Solidària at the Universitat Autònoma de Barcelona to implement a project with primary school children in which they have co-created accessible 360º videos.

This presentation will focus on such project, which has involved 6 schools close to UAB and a total of 66 children in a vulnerable situation. The project has taken place during the academic year 2021-2022 and the final results are already available. The children have co-created a 360° interactive experience video with the authoring tool Fader and have added different accessibility layers (audio description, subtitling, sign language interpreting).

We will describe the project design, the implementation –in which UAB volunteers have taken part—and the main results, which were presented in a dedicated visit of the children at UAB on May 18th, 2022.

Results show how co-creation in 360° is a motivating process that fosters social inclusion and also demonstrate how an increased awareness about accessibility has grown in the participants.

More information about the project can be found on the following links:

Anna Matamala (presenter), PhD, is a Full Professor at the Universitat Autònoma de Barcelona. Leader of Transmedia Catalonia research group, Anna Matamala has participated (DTV4ALL, ADLAB, HBB4ALL, ACT, ADLAB PRO, IMAC, TRACTION) and led (AVT-LP, ALST, VIW, NEA, EASIT, RAD) many funded projects on audiovisual translation and media accessibility. She is currently involved in the European project Mediaverse and is leading the network AccessCat.

Víctor García is a member of the TransMedia Catalonia research group at Universitat Autònoma de Barcelona. Victor is the audio describer at Liceu Opera Barcelona and teaches audio description at the School of Audiovisual Media (EMAV), Barcelona. He collaborates in European H2020 projects, with a focus on ICT, accessibility and social inclusion.

Mariona González-Sordé holds a BA in Translation and Interpreting from the Pompeu Fabra University (UPF), a BA in Applied Languages from the same university, and a MA in Audiovisual Translation from the Autonomous University of Barcelona (UAB). Before joining UAB, she worked as a subtitler and as a translation project coordinator. She is currently a PhD student at UAB, working on Easy Language and media accessibility. The aim of her studies is to empirically evaluate the recommendations in Easy Language guidelines. Her research is backed up by a scholarship linked to the Mediaverse project.

Accessible filmmaking training: a competency-based curriculum to train filmmakers in media accessibility

Florencia Fascioli Álvarez

The visibility that translation and media accessibility have on Film Studies and on audiovisual training seems to be limited compared to other areas of impact on films (Johnston, 2020). The Accessible Filmmaking (AFM) approach proposes a proactive and collaborative working model for the creation of translated and accessible films, with a flexible training proposal that aim to connect filmmakers with translators and media accessibility experts. For filmmakers, this includes specific content like the characteristics of the end users of translated and accessible films, the history and trends of media accessibility and audiovisual translation, the theory and practice of some modalities focusing on workflows, stakeholders, software and quality guidelines and standards, the AFM rationale and its practical application, and the impact of translation and media accessibility on films (Romero-Fresco, 2019). But what are the competencies of an accessible filmmaker? What curriculum design allows filmmakers to learn about translation and accessibility, and connect it with their usual creative, experimental and reflective practices?

This presentation proposes a competency-based curriculum aimed at training filmmakers in the AFM approach, with special emphasis on learning outcomes, contents, and methodologies necessary to bring filmmakers closer to translation and accessibility. The competency design is based on the results obtained in a set of 25 semi-structured interviews with film students and professionals in Uruguay (Fascioli-Álvarez, forthcoming), as well as the analysis of the training programs of all audiovisual, translation and media accessibility careers in the country. The curriculum has a structure of modules and units, and is drawn on four main competences related to, firstly, the ability to anticipate, that is, to plan, manage, evaluate, and promote translation and accessibility from the early stages of an audiovisual project. Secondly, the collaboration with other professionals and the degree of involvement within the accessibility and translation of a film. Third, the capacity to (co)create film for and with others, without losing sight of the validation stage within an iterative process with end users. And, finally, the potentiality for artistic creation and experimentation on film and media access. The presentation also outlines the preliminary results of the application of this curriculum in a training course for filmmakers within a film festival in Uruguay. This proposal is underpinned by principles in training and pedagogy used in Audiovisual Translation Studies, but also in Film Studies, Disability Studies and the so-called Accessibility Pedagogy (Greco, 2019).

Fascioli-Álvarez, F. (forthcoming). Accessible Filmmakers: Towards A Professional Profile. Journal of Audiovisual Translation, Special Issue.

Greco, G. M. (2019). Towards a pedagogy of accessibility: The need for critical learning spaces in media accessibility education and training. Linguistica Antverpiensia, New Series: Themes in Translation Studies, 18, 23–46.

Johnston, K. (2020). The look of the con: eleven thoughts on the historical absence of subtitles in film analysis. Perspectives, 28(6), 1-10.

Romero-Fresco, P. (2019). Training in Accessible Filmmaking. Linguistica Antverpiensia, New Series: Themes in Translation Studies, 14. https://lanstts.uantwerpen.be/index.php/LANS-TTS/article/view/513

Florencia Fascioli Álvarez (presenter) is a PhD student at Universidade de Vigo. She holds a master’s degree in Audiovisual Translation from the Universitat Autònoma de Barcelona and a Communication degree from the Universidad Católica del Uruguay. She has been working as a freelance audiovisual editor since 2002 as well as a cinema and theatre audio describer in Uruguay and other countries in Latin America since 2010. Her research focuses on audiovisual communication and media accessibility, with a special interest in accessible filmmaking (AFM) and audio description. Currently, she works as a professor of audiovisual production in the career of Communication, and as the Audiovisual Accessibility Program coordinator at the Humanities and Communication Department in the Universidad Católica del Uruguay. She is a member of the Interdisciplinary Nucleus on Accessibility and Communication at Universidad de la República (Uruguay).

AVT3: audiovisual translation technologies training

Verónica Arnáiz-Uzquiza and Paula Igareda

The emergence of new profiles in the language industry, together with the irruption of new forms of communication and content production, are but some of the challenges in the curriculum of professionals-to-be.The cultural, technological and scientific turns in society have always urged professionals to meet the needs of social demands, trying to bridge the gap between the new professional niches and professionals’ training (Watson, Roulstone, & Thomas, 2019). The imposing -engulfing- role of technology is giving rise to new products, needs, tools and solutions T&I professionals have to deal with, forcing academia to keep pace in such a changing scenario. Although already present in the different T&I curricula across Europe, “technologies” and “tools'' have been -and still are- an uneven part of the training process of professionals-to-be, often restricted to specific courses -such as IT Tools, CAT Tools, and more recently Machine Translation and Post Editing-, or “technological fields” such as Localisation, Audiovisual Translation and Media Accessibility in the last years (Álvarez-Álvarez & Arnáiz-Uzquiza, 2017). In this context, moved by the increasingly important role of tools and apps in the performance of translators, different training solutions are arising to help professionals satisfy the needs of the market: Higher education institutions are adapting to the new profiles by updating their curricula; postgraduate training programmes are being designed to meet the needs of the market; and reskilling courses are increasingly being offered by private companies and associations trying to help professionals adapt to a continuously changing market. Such is the case of Audiovisual Translation (AVT) and Media Accessibility (MA). Aware of their technological dependence, both training and practice have relied on essential tools for translators' performance (Bogucki & Deckert, 2020; Chaume, 2013; Chaume, 2018; Greco, 2019; Matamala & Orero, 2018; O’Hagan, 2019), and have more recently added new tools that seek to make the most of the profession, recently climbing to the cloud (Bolaños-García-Escribano, Díaz-Cintas & Massida, 2021). But, still, there seems to be a somewhat controversial divide in the role of technologies, making of the not-new, but growing presence of Machine Translation and Postediting in the AVT and MA markets something extended in industry, but not that present in the T&I training curricula yet. Leaving controversy aside, this work presents an overview on the current scene of AVT and MA in T&I training across Europe, covering the most extended formal and non formal academic programs and training modalities. Focusing on the technological component of the different curricula and content programs, the work will delve into the profile of the tools and technologies professionals and professionals-to-be are currently being trained in: from AVT “translation-only” tools, speech-to-text systems, CAT tools, MTPE, or the different tools, apps, and resources nowadays used by the language industries.

Verónica Arnáiz-Uzquiza (presenter), PhD, is an assistant professor at the Faculty of Translation and Interpreting at the University of Valladolid (Spain). Her research interests include translation training, audiovisual translation, and accessibility -especially subtitling for the deaf and hard of hearing-, and has participated in many national and international funded projects on these research fields. She coordinates the research group "Intersemiotics, Translation and New Technologies", and the European Research Project "FOIL-Online Training for the Language Industries". She is a member of the European Association for Studies in Screen Translation.

Paula Igareda (presenter) holds a degree in German Philology from the Universidad de Salamanca and a PhD in Audiovisual Translation from the Universitat Pompeu Fabra. She currently is a lecturer at the Universitat de Vic-Universitat Central de Catalunya in the Faculty of Education, Translation, Sport and Psychology. She has experience as a professional translator since 2005 and has participated in several European projects related to equal opportunities and the integration of people at risk of exclusion. Her main lines of research are Audiovisual translation, Media accessibility and translation and comic studies (she is currently doing a MA in Comic and Education at the Universitat de València).

12.00 - 13.30: Lunch

13.30 - 15.00: Panel 3 - Chances and challenges of multimodality and translanguaging for inclusive communication in higher/tertiary education – an international and interdisciplinary perspective

Maija Hirvonen, Yanou van Gauwbergen, Irene Hermosa-Ramírez

This panel discussion is organized by the international Network on Inclusion and Inclusive Communication in Tertiary Education (NIICTE, launched in 2021) whose goal is to support partnerships that link research in inclusion in the higher education context. Inclusion in educational settings encompasses teaching arrangements and learning materials, didactic approaches as well as multimodal and translanguaging practices for participants with dis/abilities and specific needs. This refers to participants using specific communication methods, such as sign languages or audio-description and braille, but it also refers to the adaptation of (language) behaviors for students with mental diversity. The potential and the multimodal/communicative resources of these communicative practices for learning and innovation have been understudied in both research and practice.

The increase in digital and multimodal teaching brings along various challenges and chances for inclusion. Neurodiverse individuals may experience difficulty filtering relevant stimuli during information processing, interpretation, and social interaction due to stimulus overload. On the other hand, studies on multimodality have established that combining different modalities according to the Universal Design for Learning (UDL), such as visualization, body movement, and written or oral communication, can support mutual understanding in an interaction involving participants with special abilities and needs.

In the light of these chances and challenges, the panel discusses with the audience the following topics:

How can and should we bring Accessibility and Design for All into Higher Education curricula? Training and education on these subjects are currently on the rise across society (from companies to universities). Developing and testing guidelines and recommendations will stimulate innovative learning and teaching practices to tackle social inclusion of persons with disabilities.
What is the role of human and technological agents in enabling inclusion? The use of automatic translation and transcription (live subtitles) in the classroom, for instance, can be taken to replace human translators and interpreters, but what are the consequences?
How can we improve the inclusion of students with mental health issues (e.g., anxiety or depression), which fall under the category of invisible dis/abilities, and what multimodal practices can be applied in teaching/learning contexts? The digitalization of teaching at universities offers both opportunities (i.e., digital recordings) and risks due to social acceleration for these groups.
How can websites of higher education institutions support or prevent inclusion? A look at websites of higher education institutions shows that the design of their interactivity can be challenging due to incoherent terminology, unclear positioning of actors, and lack of intersectionality for the exchange with heterogeneous students. It all becomes more difficult when websites are (automatically) translated for international students with dis/abilities.

Abrahamson, Dor; Flood, Virginia J.; Miele, Joshua A.; Siu, Yue-Ting (2019). Enactivism and ethnomethodological conversation analysis as tools for expanding Universal Design for Learning: the case of visually impaired mathematics students. ZDM - Mathematics Education 51(2).

ATHENA. https://www.edf-feph.org/projects/athena-bringing-accessibility-and-design-for-all-into-higher-education-curricula/

Danforth, Scot, Gabel, Susan L. (2017). Vital Questions Facing Disability Studies in Education. New York (US): Peter Lang.

Holmström, Ingela; Schönström, Krister (2018). Deaf lecturers‘ translanguaging in a higher education setting. A multimodal multilingual perspective. Applied Linguistics Review 9(19).

Koponen, Maarit & Tuominen, Tiina & Hirvonen, Maija & Vitikainen, Kaisa & Tiittula, Liisa. (2021). User perspectives on developing technology-assisted access services in public broadcasting. Bridge: Trends and Traditions in Translation and Interpreting Studies 1(2).

Media Accessibility Platform. https://mapaccess.uab.cat

Moriña, Anabel; Sandoval, Marta & Carnerero, Fuensanta (2020). Higher education inclusivity: when the disability enriches the university. Higher Education Research & Development 39(6).

NIICTE. https://research.tuni.fi/niicte/

Rosa, Hartmut (2013). Social Acceleration: A New Theory of Modernity. New York Chichester, West Sussex: Columbia University Press.

Sowińska, Agnieszka. (2022). “They say it’s because I’m migrainous. . .” Contested identities of students with invisible disabilities in medical consultations. Discourse Studies 24(3).

Maija Hirvonen (convener of the panel), Tampere University, Faculty of Information Technology and Communication, Languages Unit & Tampere Accessibility Unit, Network of Inclusion and Inclusive Communication in Tertiary Education. Research focus: Accessibility of communication and culture (especially Audio Description), Asymmetric Interaction (between blind and sighted people), Distributed Cognition.

Yanou van Gauwbergen, University of Antwerp, Department of Applied Linguistics, Translation and Interpreting, Translation, Interpreting and Intercultural Studies Research Group. Research focus: Accessibility through interlingual live subtitling in higher education.

Irene Hermosa-Ramírez, Autonomous University of Barcelona, Department of Translation, Interpreting and East Asian Studies, TransMedia Catalonia research group, ATHENA project (Bringing Accessibility and Design for All into Higher Education Curricula). Research focus: Accessibility and universal design in higher education curricula and syllabi.

15.00 - 15.30: Break

15.30 - 17.00: Session 12 - Research pitches AVT

A study of translation policy in the subtitling industry in China: an actor-network approach

Lu Yan

The sociological research in Translation Studies (TS) continues to underpin agential networks involved in different translation processes at diverse global locations. However, current knowledge remains scarce when it comes to the AVT praxis taking place outside European and Anglophone contexts. This study aims to fill the current gaps by focusing on the subtitling practice in China. Applying Latour’s (1996) Actor-Network Theory (ANT), the study will underpin the policymaking and translating actors and illuminate the relations between them in the production process of professional and fan-based subtitling in contemporary China on the one hand; and shed light on how the actors involved in subtitling practice negotiate their positionalities in dealing with policy-related issues on the other hand.

By doing so, the research aims to answer the following questions:

What kind of evidence of translation policies can be detected in the subtitle production process in the Chinese context?
Who and what are involved in subtitling and relevant policymaking, and how do they interact?
How are policies formulated by policymakers? How are they complied with or contested by subtitling professionals and fansubbers?
What are the similarities and differences between these actors in professional and fan-based contexts?

Ethnographic work will be conducted with the local authorities, commercial service providers and fansubbing groups to investigate how policies are complied with and negotiated across the spectrum of subtitling agents and policymakers. Multiple methods will be adopted for quantitative and qualitative data collection, such as: 1) administering questionnaires and interviews with professionals, fansubbers, members of management and policymakers, 2) collecting documents encompassing text and visual data, which include policy documents, guidelines, media reports and any other available documentation, and 3) making participant observation, ranging from field notes (my thoughts, comments and experiences) to photographs, audio and video recordings, which are the traces of human and nonhuman actors. By assembling distinctive actor-networks of policymaking and subtitling agents, the research will deliver a comprehensive survey of the policies that modulate the choices and strategies adopted by the multiple subtitling agents at various levels in China.

Latour, B. (1996) ‘On Actor-network Theory: A few clarifications’, Soziale Welt 47(4): 369–381.

Lu Yan, a PhD candidate in Translation Studies at the University of Geneva since 2021, has always been fascinated by the complex relationship between heterogeneous actors that underpin in the production process of translation, and the ways in which these actors negotiate their positionalities in dealing with decision-making issues. Her research, directed by Prof. Alexander KÜNZLI at the University of Geneva and Prof. Dingkun WANG at the University of Hong Kong, focuses on translation policy in China’s subtitling industry. Applying Latour’s Actor-Network Theory (ANT), it could provide sociological insights into the production process of audiovisual translation (AVT) based on ethnographically empirical data. Her primary research interests are in translation technology and audiovisual translation, specifically subtitling.

Using social media to understand subtitlers’ concerns

Tinka Stössel

For many of us, browsing social media is something that we do in our spare time and rather mindlessly before going to sleep or on the way to work. While much of the content is geared towards this down time, it is also possible to find information that can be highly beneficial for research, especially for those who are new to the industry. In my doctoral dissertation, I am focusing on the sustainability of Netflix’s subtitling process from the subtitlers’ perspective. In recent years, a growing number of subtitlers have voiced their dissatisfaction with various aspects in their work, such as rates, their interaction with other agents in the process, and the input materials provided by language service providers, to name a few. With an increase of suboptimal conditions, many even admit to wanting to quit, thus suggesting that the current situation cannot continue in the future, i.e., is not sustainable. My research aims to identify the key concerns of subtitlers affecting the sustainability of the current subtitling process and whether these are also found in Netflix’s process. To do this, my methods of gathering data will include a questionnaire and interviews where I ask subtitlers with German as a target language about their experience working (indirectly) with Netflix and whether they consider the current process to be sustainable. As a first step, however, I conducted a social media sweep to build upon my literature review and identify any under-researched topics that should be included in the questionnaire and interviews. This method of data collection included analyzing posts on Facebook groups dedicated to subtitling. While some subtitlers’ concerns, such as low rates, guidelines, and the quality of the end-product are well-known throughout the industry, other aspects such as non-disclosure agreements, royalties, and the self-assignment scheduling system have not – to my current knowledge - had as much attention in the literature. As valuable as this information is, it is important to be cautious with social media data. Because of the nature of the internet and posts being based on personal experiences, opinions can vary significantly and there is always the possibility of posts being deleted or amended. The question of whether and how a post can be cited is also controversial as this is still a relatively new field of data collection. For the researcher, it can also be mentally and emotionally draining to read through and analyze posts for extended periods of time, especially when posts consist of heated discussions. At this point in my research, I am still reviewing the social media data, which will form the basis for the questionnaire that I am hoping to distribute by the fall of 2023.

Tinka Stössel is a research assistant and PhD candidate at the University of Geneva. She has a BA in modern languages (French and German) from Texas A&M University and an MA in cultural and cognitive linguistics from the LMU Munich and is currently preparing a dissertation in audiovisual translation. She previously worked as a translator in a Munich-based patent law firm and assisted in two research projects during her MA studies: 1) transcribing audio content for the International Corpus of English and 2) evaluating a French-language program in preschools as part of the Franco-German Elysée-Kitas project. Since 2021, she is a research assistant in the German Unit of the Faculty of Translation and Interpreting and an editorial assistant for the translation studies journal Parallèles. For her dissertation, she is focusing on the sustainability of the subtitling process in streaming with a Netflix case study.

Machine translation in video games - a pilot study on post-editing subtitles

Judith Brenner

So far, not much academic research has been conducted about video game localization and the use of translation technology, and knowledge about this field is mainly based on industry use cases and anecdotal evidence. Now with machine translation added to the mix of available translation technology, selecting the right tools for an efficient translation process becomes an ever more complex task. By now, several game developers and publishers have adopted post-editing (PE) of machine translation (MT) suggestions as one possible workflow when translating video games, but still many game translators say that they perceive MTPE to be slower than translating from scratch. They also think it hinders them in their creative freedom and that it does not result in the same high-quality creative translations as without MTPE. My doctoral research aims to test this perception and to find MTPE methods that are faster, more satisfactory, and help achieve expected quality levels, and thus are better accepted by game translators for their creative work.

In my pilot study to be conducted during 2023, I will compare three different MTPE methods when translating subtitles for YouTube videos about video games published by a cooperating game developer based in Finland. The different MTPE methods are as follows: 1) traditional MTPE, where all sentences of a text are either pre-populated with high fuzzy matches or pre-translated by NMT; 2) NMT-assisted translation, where the target segments are not pre-populated so that translators can more freely choose for every sentence to either use an NMT suggestion, reuse (and edit) a TM match, or translate from scratch; and 3) adaptive NMT that adapts almost in real time to previous post-edits. Analyses of the three different MTPE methods will be quantitative as well as qualitative to uncover their impact on translation process, translator satisfaction, and translation quality. The translation process will be analyzed through keystroke logs that record edit times to measure the productivity of the three different methods and pauses to infer cognitive effort. Eye-tracking will add further insights on the cognitive effort by the translators under the three different methods to link the translation process to the translator satisfaction that will further be investigated through retrospective surveys. Lastly, quality evaluations of the final translations will examine the level of quality achieved through the three different MTPE methods. Additional analysis will compare productivity, satisfaction, and quality of the individual translators using the three different methods.

Insights into the productivity of and satisfaction with using these different MTPE methods and the resulting translation quality should support game translators in finding their preferred MTPE method so that they can achieve the speed increase that developers and publishers desire from using MT.

At Media For All 10 , I will present a project plan for this pilot study to gather input from attendees about how to improve my study design both for the pilot study as well as for future experiments involving texts appearing on screen within video games rather than video subtitles.

Judith Brenner is a doctoral researcher at the University of Eastern Finland and Cologne University of Applied Sciences, where she is investigating different modes of post-editing machine translation output in video game translation. Before turning to research, Judith worked as Localization Editor for German at the game developer Blizzard Entertainment in France. She also has more than ten years' experience as a freelance translator and editor for PC hardware, software and video games, and holds a first degree in Translation Studies and a master's degree in Terminology and Language Technology.

Non-literal language use in machine and human translation for dubbing

Milena Chaîne

The goal of this PhD project is to investigate the ways in which human translators and machine translation systems handle non-literal language (e.g. idioms, metaphor, jokes, conventional indirectness, or conversational implicatures) when translating scripts for commercial lip-sync dubbing of films. Non-literal language features play a central role in film as they drive forward plot and character development. However, they are problematic for existing machine translation (MT) approaches for a variety of reasons. For example, MT systems may struggle to infer culture-specific implied meanings, creative or linguistically innovative language use, or the intended meaning of an utterance relying on the extended linguistic or visual context in which it is embedded. MT approaches to translation depend on linguistic surface structures to identify language patterns and therefore struggle with identifying pragmatic, context-dependent meanings. Human translators, on the other hand, may not struggle to recognise non-literal language in a source text, but they may have difficulties recreating it in equivalent and lip-synchronous ways, while taking dubbing constraints into account, in the target language. This PhD project, in partnership with an industry leader in media localisation, seeks to integrate the distinct qualities of MT and human translators to identify linguistic, textual and visual characteristics of non-literal language in film within the context of dubbing. These could then be expressed in a computer-parsable way to help identify areas in which MT approaches struggle and where human translators should focus their efforts. Research approaches to human translation and MT are often isolated from one another and rely on two distinct views of translation: translation as a culture-based creative process in translation studies, and translation as a computational-statistical procedure in MT research. This project aims to bridge the gap between these approaches to translation in order to offer a solution to a commonly identified problem in professional audiovisual translation (AVT) practice, and better understand the role of the human translator in relation to MT and human-machine translation systems for dubbing.

The role of human translators in the translation process, as well as their relationship to MT and MT systems is an essential part of this research project. As a result, to help define the scope and direction of further research in the project, we are planning to conduct unstructured, open interviews with professional translators, transcribers and engineers working on commercial dubbing pipelines. This will provide us with valuable qualitative data on their perception of the translation issues related to non-literal language features, and of MT approaches to translation for dubbing. It will also inform the design of a following observation study of the behaviour of human dubbing translators as they translate different types of context-embedded non-literal language. This study will include self-report methods like think aloud protocols and direct measures like eye-tracking, keylogging, and screen recording. Overall, these studies will provide evidence of translation decision-making procedures and provide a human processing perspective on a formal description of non-literal language, as well as the relationship between human agents and MT systems in the field of translation.

Milena Chaîne (presenter) is a second-year PhD student in the School of Languages and Cultures at the University of Sheffield (United Kingdom), working on non-literal language use in human and machine translation for dubbing. She is supervised by Dr Nicole Baumgarten and Dr Lena Hamaidia. She previously completed a master's degree in Natural Language Processing at the Inalco (Paris, France).

17.00 - 18.30: Session 13 - Research pitches Accessibility

Accessibility and design for all in higher education curricula: the athena project

Anna Matamala, Irene Hermosa-Ramírez, Carme Mangiron

Accessibility and Design for All are central for the full participation of persons with disabilities in all societal domains, from education to work, culture or entertainment, to name just a few. Professionals from a wide range of areas should learn about guidelines, standards, techniques and tools that facilitate access to persons with disabilities, their communities and society as a whole. However, Accessibility and Design for All are still not part of all higher education curricula. The ATHENA project aims to overcome this gap by developing a set of recommendations on how to integrate Accessibility and Design for All into higher education. The project has been funded under the Erasmus + call. It will run for 30 months (ref. code 101089469), under the leadership of EDF (European Disability Forum).

This presentation will focus on: a) general presentation of the project, and b) preliminary results of the first six months of the project. We hope to engage the audience in a discussion on how Accessibility and Design for All could be integrated in a wide variety of degrees.

Irene Hermosa-Ramírez, PhD in Translation and Intercultural Studies (UAB), is a postdoctoral researcher working for the Erasmus+ ATHENA project and a member of the TransMedia Catalonia research group. Among her latest publications is the article “Physiological instruments meet mixed methods in Media Accessibility”. Irene is the secretary of the Catalan Association for the Promotion of Accessibility.

Carme Mangiron, PhD, is a lecturer and a member of the research group TransMedia Catalonia at UAB. She is the Chair of the MA in Audiovisual Translation. Her research interests include game localization, game accessibility AVT and accessibility to the media. She has published extensively and participated in several research projects. She is co-author of Game Localization: Translating for the Global Digital Entertainment Industry (O’Hagan and Mangiron, 2013).

Audio description of enjaulado (1997) and vinil verde (2004): intersemiotic translation of formal elements

Marcella Wiffler Stefanini

We will present an excerpt from and ongoing doctoral research that aims to study the intersemiotic character of the translation involved in an audio description (AD), in particular the translation of formal elements of audiovisual texts. For this, in addition to presenting the theoretical framework on intersemiotic translation and, more specifically, on the relationship between image and writing, an AD script will be made for two short films by the Brazilian director Kleber Mendonça Filho: “Enjaulado” (1997), written and directed by Kleber Mendonça Filho, and “Vinil verde” (2004), whose screenplay is signed by the Brazilian director and the Ukrainian filmmaker Bohdana Smyrnov. We propose to prepare the script with two visually impaired audio describers, designated as consultants in AD, to also study how the construction of meanings of formal imagery elements takes place in this collaborative process between sighted person and visually impaired person. The methodology will therefore involve the documentation of the process of writing the scripts and the interaction with the AD consultants, so that, even during the process, it may be possible to establish relationships between the practice and the theory, taking as referential the semiotics of Peirce, along with the studies published by Plaza (1987), Metz (1968) and Flusser (1972). In addition to these, we will adopt another theoretical framework to analyze the two selected shorts films, especially those theories related to the study of the “Árido movie”, since, in the wake of Candido (1985), we defend the importance of considering the external factors when interpreting a work of art, especially regarding its internal factors, that is, the importance of context for understanding the formal aspects of a work of art. In this sense, while writing the scripts, we will also take into account researches developed around the study of "Árido movie", such as Nogueira (2009), Paiva (2016, 2020), and Nagib (2014, 2020). This artistic movement, which takes place in Pernambuco, began in the 1990s and is characterized by the intermediality between cinema and other media, especially music. Alongside, we will consider other specific theories that have been proved to be relevant, such as the studies of Elias (2014) on photo film and Elsaesser (2018) on the aesthetics of obsolescence. We expect to demonstrate the importance of working with an AD consultant and knowing about film aesthetics (Alves; Araújo 2016) and intersemiotic translation for the production of an AD script that promotes accessibility to the artistic content of an audiovisual work.

Marcella Wiffler Stefanini is currently a PhD student at the Post-Graduate Program in Applied Linguistics (PPG-LA) at the State University of Campinas (Unicamp), within the line of research Languages, Transculturality and Translation. She holds a master's degree in Applied Linguistics (2020) and a degree in Letters (2017), also from Unicamp. At the PPG-LA, she participates in the research group "E por falar em tradução", which develops a project focused on translating into portuguese texts related to the Translation Studies, published periodically under the title "As diversas faces da tradução na contemporaneidade". Since the graduation, she has studied audio description (AD) from the perspective of the Translation Studies and today, in her PhD research, she focuses on the study of AD as an intersemiotic translation and its possible associations with filme language. She is interested in topics related to the Translation Studies, Audiovisual Translation, Audio Description and Cinema.

Accessible cues: a novel mode to improve access to audiovisual narratives for cognitively diverse audiences

Andreea Deleanu, Kim Starr, Constantin Orasan, Sabine Braun

Media Accessibility (MA) has come a long way in improving the lives of people with sensory disabilities. Driven by legal requirements and technological advances, the field has demonstrated that disabilities do not need to hinder cultural participation (European Commission, 2015; Ofcom, 2017). As a case in point, Audio Description (AD) is now a well-established means to improve access for people with visual impairment. However, it has also proven beneficial for other audiences, e.g. in learning environments (Cozendey & da Piedade Costa, 2016), as it facilitates knowledge acquisition and guides attention (Krejtz, et al., 2012; Walczak, 2016). Furthermore, traditional AD has been adapted to support emotion recognition (Starr & Braun, 2021) for high-functioning autistic children, suggesting that its repurposing can overcome more than just sensory barriers and support Cognitive Accessibility (Johansson, 2016). To date, training for the adaptation of traditional AVT modes has been mainly investigated within the EASIT project*. However, the question of how to reduce cognitive barriers and achieve social inclusion and cultural participation for people with diverse cognitive abilities, as required by the UN Convention on the Rights of Persons with Disabilities (UNCRPD, 2006), needs to be further explored.

This presentation introduces a research project which addresses this requirement by focusing on people with diverse cognitive abilities, so far excluded from fully accessing audiovisual storyworld narrative (Herman, 2002). This is achieved through the repurposing of AD and the use of Easy-to-Understand (E2U) language. The expected outcome is the creation of a novel access mode: Accessible Audio Cues.

This project starts from two premises. First, that processing complex audiovisual narratives such as films and TV productions can prove cognitively challenging (Schlickers, 2009); second, that an individual’s processing ability is both hindered and facilitated by their unique processing functions, world knowledge, cognitive and language-related abilities (APA, 2013). At the same time, studies which have applied theories of understanding and narratology to the processing of (audiovisual) narrative appear to support the assumption that the cognitive effort required in this process can be reduced by strategies such as explicitating complex inferences and providing support in terms of frames of reference, focalization, and context-relevant information (e.g., Kruger, 2012; Braun, 2016; Vercauteren, 2021 on audiovisual narrative; Merritt & Liles, 1987; Wolman, 1991; Loth, et al., 2011 on processing of cognitively diverse individuals).

Based on this work, the current project has conducted a comparative analysis of text-based E2U practices and guidelines, and has analysed two corpora of simplified articles to identify relevant linguistic patterns in the simplification of written text that can inform the development of simplification strategies for the verbal description of audiovisual narrative. The next step will be to develop simplified descriptions for audiovisual material, validate them with experienced E2U practitioners and pilot them in a reception study to gauge the extent to which they can improve cognitive access to, comprehension, and enjoyment of audiovisual fiction. The presentation will outline the project’s background, methodological approach and initial key findings.

EASIT, Easy access for social inclusion training, is a project funded by the Erasmus+ Strategic Partnerships programme (2018-1-ES01-KA203-05275).

APA, A. P. A., 2013. Diagnostic and statistical manual of mental disorders. 5th ed. ed. Arlington: APA. Braun, S., 2016. The importance of being relevant? A cognitive-pragmatic framework for conceptualising audiovisual translation. Target, 28(2), pp. 302-313.

Cozendey, S. & da Piedade Costa, M., 2016. The Audio Description as a physics teaching tool. Journal of Reserach in Special Educational Needs, Volume 16, pp. 1031-1034.

European Commission, 2015. The European Accessibility Act. [Online] Available at: https://ec.europa.eu/social/main.jsp?catId=1202

Herman, D., 2002. Story Logic: Problems and possibilities of narrative. s.l.:University of Nebraska Press.

Johansson, S., 2016. Towards a framework to understand mental and cognitive accessibility in a digital context. [Online] Available at: http://kth.diva-portal.org/smash/record.jsf?pid=diva2%3A908641&dswid=4992

Krejtz, I., Szarkowska, A., Krejtz, K. & Duchowski, A., 2012. Audio description as an aural guide of children's visual attention: evidence from an eye-tracking study. s.l., ETRA '12: Proceedings of the Symposium on Eye Tracking Research and Applications, pp. 99-106.

Kruger, J.-L., 2012. Making meaning in AVT: eye tracking and viewer construction of narrative. Perspectives: studies in Translatology, 20(1), pp. 67-86.

Loth, E., Gómez, J. C. & Happé, F., 2011. Do high-functioning people with autism spectrum disorder spontaneously use event knowledge to selectively attend to and remember context-relevant aspects in scenes?. J Autism Dev Disord, 41(7), pp. 945-961.

Maaß, C., 2020. Easy Language – Plain Language – Easy Language Plus. Balancing Comprehensibility and Acceptability. Berlin: Frank&Timme.

Merritt, D. & Liles, B., 1987. Story grammar ability in children with and without language disorder: story generation, story retelling, and story comprehension. Journal of Speech, Language and Hearing Research, 30(4), pp. 539-552.

Ofcom, 2017. Ofcom Code on Television Access Services. [Online] Available at: https://www.ofcom.org.uk/__data/assets/pdf_file/0020/97040/Access- service-code-Jan-2017.pdf.

Schlickers, S., 2009. Focalization, Ocularization and Auricularization in Film and Literature. In: W. S. a. J. S. Peter Hühn, ed. Point of View, Perspective, and Focalization: Modeling Mediation in Narrative. Berlin: De Gruyter, pp. 243-258.

Starr, K. & Braun, S., 2021. Audio description 2.0 Re-versioning audiovisual accessibility to assist emotion recognition. In: Innovation in Audio Description Research. London: Routlege, pp. 97-120.

UNCRPD, 2006. United Nations, Convention on the Rights of Persons with Disabilities. [Online] Available at: https://www.un.org/development/desa/disabilities/convention-on-the-rights-of-persons-with-disabilities.html

Vercauteren, G., 2021. Insights From Mental Model Theory and Cognitive Narratology as a Tool for Content Selection in Audio Description. Journal of Audiovisual Translation, 4(3), pp. 6-24.

Walczak, A., 2016. Foreign Language Class with Audio Description: A Case Study. In: A. O. P. Matamala, ed. Researching Audio Description. Palgrave Studies in Translating and Interpreting.. London: Palgrave Macmillan, pp. 187-204.

Wolman, C., 1991. Sensitivity to causal cohesion in stories by children with mild mental retardation, children with learning disabilities and children without disabilities. The Journal of Special Education, Volume 25, pp. 135-154.

Andreea Deleanu (presenter) is a PhD Candidate at the University of Surrey, UK, specialising in Media Accessibility (MA). Her project focuses on repurposing audio access modes for cognitively diverse audiences to fully access audiovisual narratives. The ensuing mode will also reduce the effort stemming from the decoding of additional verbal material through the use of Easy-to-Understand (E2U) audio cues. Andreea holds an MA in Conference Interpreting from the University of Trieste (Italy) and has specialised as a healthcare interpreter. She was part of the EASIT project, working on E2U and Audio Description (Unit 3B).

Kim Starr, PhD, is a Research Fellow in the Centre for Translation Studies (CTS) at the University of Surrey (UK). She previously worked in commercial television, and has degrees in politics and law (BSc.Econ., QMUL), journalism (MA, Westminster) and audiovisual translation (MA, University of Surrey). Awarded a doctoral scholarship by the Arts and Humanities Research Council/TECHNE, she completed her PhD in audio description for cognitive diversity in December 2017. For the following three years she worked on the EU Horizon 2020 project ‘Methods for Managing Audiovisual Data’ (MeMAD). She is currently employed as part of the Expanding Excellence in England (E3) programme at the CTS, researching human-machine interaction in the field of automated video description, cognitive accessibility in cultural contexts, and audio description for the non-blind. Kim co-edited and co-authored the IATIS yearbook, ‘Innovation in Audio Description Research’ (Braun & Starr, 2021). In 2022, she was awarded the Fulbright-Smithsonian Scholarship.

Constantin Orasan is Professor of Language and Translation Technologies at the Centre of Translation Studies (CTS), University of Surrey (UK). He has over 20 years experience of working in the fields of Natural Language Processing (NLP), Artificial Intelligence and Machine Learning for language processing. His research interests are largely focused on facilitating information access and include translation technology, text simplification, sentiment analysis, question answering, text summarisation, and anaphora and coreference resolution. His recent projects focused on using NLP to facilitate access to justice.

Sabine Braun is Professor of Translation Studies, Director of the Centre for Translation Studies at the University of Surrey, and a Co-Director of Surrey’s Institute for People-Centred AI. Her research explores human-machine interaction and integration in translation and interpreting, especially to improve access to critical information, media content and vital public services. For over 10 years, she has led a research programme that investigates the delivery of interpreting services via video link to improve language access in the public sector. In addition, she is investigating the feasibility of (semi-automating) audio description (video-to-text translation) to improve media access for diverse audiences.

Time for a community-based participatory research turn in media accessibility?

Irene Hermosa-Ramírez

The user-centred research turn in Media Accessibility (MA) (Greco, 2018) is here to stay, as illustrated by the very themes of the Media for All 10 conference. Concurrently, the involvement of users in the creation and design of accessible services has been advocated for in the form of participatory accessibility for the scenic arts (Di Giovanni, 2018), poietic design (Greco, 2019) and accessible filmmaking (Romero-Fresco, 2019). Yet, user-led research in MA has been scarcer, and it has mostly stemmed from researchers outside the field of Translation Studies (see, for instance, Cavallo, 2015; Kleege, 2016; and Thompson in Chottin and Thompson, 2021). This paper introduces the Community-Based Participatory Research (CBPR) approach as a viable framework to be applied in MA. It does so by gathering two arguments advocating for the viability of CBPR, along with two challenges to be anticipated.

As a general framework, the logic behind CBPR includes “forming research partnerships with non-academic stakeholders to develop and execute a research project based on a particular community-identified problem or issue” (Leavy, 2017, p. 224). Within CBPR, the enquiry itself is then to be translated into action: the aim is to (durably) resolve the research problem that initiated the project. A vast array of methodologies are applicable within this framework, although CBPR involves a cyclical design, as in action research (Cravo and Neves, 2007).

The first argument for the application of CBPR to MA refers to the already well-established relations between user associations and researchers in the field. Indeed, one of the main principles of CBPR refers to the collaborative partnerships and the power-sharing dynamics fostered throughout all of the research phases (Israel et al., 2008). The second argument refers to the parallel between most of our research and traditional “post-hoc” access services. The demand for involving users in the creation of accessible services or products can be closely replicated in our research itself if the CBPR framework is applied. If we advocate for the involvement of the users in the industry, the next step is to practice what we preach in academia.

Finally, out of the two main drawbacks to be expected in applying CBPR to MA, we may first anticipate ethical review committees unfamiliarised or not aligned with CBPR. Issues may arise due to the distinction between researcher and researched being blurred, the research process becoming less predictable because of its cyclical nature, and anonymity not always being guaranteed. Second, if the communities participating in the research are defined too narrowly, there is a risk of reinforcing a “ghetto effect” in MA (Greco, 2016).

Irene Hermosa-Ramírez (presenter), PhD in Translation and Intercultural Studies (UAB), is a postdoctoral researcher working for the Erasmus+ ATHENA project (Bringing Accessibility and Design for All into Higher Education Curricula) and a member of the TransMedia Catalonia research group. She is also working as a research support technician for the UNIVAC project devoted to sensory accessibility at Spanish universities at Universitat Pompeu Fabra. Among her latest publications are the articles “The hierarchisation of operatic signs through the lens of audio description” and “Physiological instruments meet mixed methods in Media Accessibility”. Irene is the secretary of the Catalan Association for the Promotion of Accessibility.