Anna Jankowska TT. asst. research professor

Research Anna Jankowska

Research team

Expertise

Audio description, audiovisual translation, translation process research

Bridging accessibility, participation and inclusion with automatic live subtitles (In4All). 01/02/2025 - 31/01/2027

Abstract

Background: Accessibility is usually understood as means of making cities more accessible for people with mobility disabilities. But accessibility encompasses also access to culture and information. Public institutions communicate with their residents and visitors through various channels. Yet, what about those who cannot access this information such as people with sensory disabilities or, as is common in the EU, people with different linguistic backgrounds? A way to provide universal access to audiovisual products is using automatic subtitling, an AI-based technology that automatically generates subtitles for pre-recorded and live audiovisual products. Aim: I will address the role of automatic subtitling in enhancing the accessibility of communication in public institutions and its potential impact on the well-being and civic engagement of citizens. Urgency: Advances in artificial intelligence have improved drastically the quality of live automatic subtitling, but we do not know how this technology impacts its users in several communicative situations, since there are no reception studies on this matter. Approach: This is an empirical and interdisciplinary study mixing ethnographic and experimental approaches. Using both qualitative and quantitative data collection tools, this is the first study investigating the interaction between live automatic subtitling and its end users. Innovation and impact: 1) Research into a new, unclaimed research avenue; 2) Interdisciplinary approach with data collection tools never used before on media accessibility; 3) Determine the impact of technology on reception of media accessibility; 4) Results will directly improve several stakeholders' lives; 5) Results will be directly applicable to professional practice, promoting knowledge-based economy

Researcher(s)

Research team(s)

Project type(s)

  • Research Project

Audio Description Expertise & Quality: Unraveling A Conceptual Yardstick (ADEQUACY). 01/11/2024 - 31/10/2026

Abstract

Ever close your eyes and listen to a movie? Millions of blind people experience this daily reality relying on audio description (AD). As a media accessibility service, AD translates visuals into an auditory experience for non-sighted audiences. But amidst the surge in audio-described content, a fundamental question persists: what defines expertise and quality in AD? While research on AD has flourished over the past 20 years within Translation Studies, the concepts of AD expertise and quality remain elusive. Alongside these knowledge gaps, what happens during the AD process is unknown. This project aims to conceptualize AD expertise and quality through a comprehensive mixed-methods study. On the qualitative front, we map perceptions of expertise and quality among sighted and non-sighted audio description professionals through DELPHI surveys and focus groups. By involving individuals with visual impairments, the project embraces a participatory approach to bridge the maker-user gap within media accessibility, where the perspectives of disabled users are often disregarded by industry figures and policymakers. On the quantitative front, the project follows the experimental approach of Cognitive Translation Studies to study the AD process and to explore whether specific behaviors of audio describers correlate with expertise and quality. In this phase, the activities of 12 audio describers will be monitored through keylogging, eye-tracking, as well as screen and voice recording.

Researcher(s)

Research team(s)

Project type(s)

  • Research Project

TAP: tapping the audio description process. 01/10/2024 - 30/09/2028

Abstract

Audio description (AD) is an accessibility service for people with vision loss. In films, it is an additional audio track which describes the visual cues necessary to understand and enjoy the film. It is a translation of images into words, also known as intersemiotic translation. There are two workflows to produce AD scripts: (1) writing - choosing the relevant visual elements and turning them into text that fits in the limited space between the dialogues and the important elements of the original soundtrack; (2) translation of existing AD scripts (e.g., English>Dutch). These workflows exist in professional practice, but research into AD processes is non-existent. The project has three aims: (1) to conceptualize the AD process model(s): (2) to conceptualize intersemiotic translation as a process; (3) to test the reactivity of think-aloud protocols (TAP) in AD process research. We conduct four experiments, two on AD-writing and two on AD-translation. We monitor the participants' work with keylogging, eye tracking, screen, voice and face recording, and interviews. Participants will be asked to think aloud in half of the tasks. This project is urgent to avoid a one-sided view of audiovisual translation and media accessibility practices as a product. It is also relevant since it elicits essential questions that product-oriented approaches cannot answer and complements existing methodological approaches in AD research.

Researcher(s)

Research team(s)

Project type(s)

  • Research Project

Accessibility in Focus: Towards a Consolidation of Accessibility Studies Across Disciplines. 01/12/2023 - 30/11/2025

Abstract

In recent decades, access and accessibility have emerged as pivotal forces reshaping a wide spectrum of human endeavours, from innovative practices through social discourse to regulatory frameworks. This transformation has propelled accessibility into the forefront of diverse research fields, catalysing a wealth of novel methodologies and conceptual frameworks that have revolutionised the foundations of these disciplines. These developments have coalesced into what Greco (2018 and 2022) identifies as Accessibility Studies—an interdisciplinary field of research dedicated to investigating accessibility processes and phenomena. Despite its promising growth, Accessibility Studies remains fragmented, with accessibility often addressed disparately within distinct academic and professional domains, including Translation Studies, Architecture, Disability Studies, Design, Information Technologies, Law, Philosophy, and others (Greco 2022). This fragmentation has resulted in a disconnected and detached landscape within the field. The objective of the present project is to lay the foundation for an EU-level consolidation of the burgeoning field of Accessibility Studies in Europe by critically mapping the current state of the art in accessibility processes and phenomena. Our project seeks to bridge the gaps and enhance cohesion by fostering a deeper understanding of accessibility within and across these diverse domains. To accomplish this, we will create an international interdisciplinary team that will collaboratively investigate accessibility from various angles in order to get a better view of its complex identity and to uncover disparities that have to be overcome to achieve a more unified and integrated perspective. This research endeavour will explore the design, implementation, and evaluation of accessibility-based and accessibility-oriented methodologies, ultimately advancing our comprehension of access and accessibility in contemporary society.

Researcher(s)

Research team(s)

Project type(s)

  • Research Project

The effect of professional experience on audio description process. 01/10/2023 - 30/09/2027

Abstract

BACKGROUND: Audiovisual content prevails in many modern societies. However, not all audiences can access the audiovisual content in its original form due to language barriers or sensory impairments (Greco, 2016). This is where audiovisual translation (AVT) and media accessibility (MA) come into play, providing access services such as e.g., subtitling, dubbing, captioning and audio description (AD). AVT&MA is the most dynamically developing area of TS (Remael, Reviers, & Vandekerckhove, 2016). While it produced a large quantity of research in less than 30 years, the majority of current AVT&MA studies are descriptive and prescriptive. Calls for more varied and in particular empirical research methods to support the AD field's development into a mature discipline (PerezGonzalez, 2014) have led to a spiked interest in experimental research. However, two crucial research gaps remain in this respect: (1) Experimental research in AVT&MA, is a fledgling area that experiences growing pains (Di Giovanni & Gambier, 2018). The predominant AVT&MA experimental research paradigm still uses offline measurement methods (e.g., questionnaires, interviews, focus groups etc.) (for a detailed overview see (Cintas & Szarkowska, 2020; Orero et al., 2018). (2) Process research approaches are almost inexistent in AVT&MA research. The field of AVT&MA remains largely focused on the product. Only a handful of studies employed the translation process research framework within AVT&MA (Gary Massey & Jud, 2020; Orrego-Carmona, 2018). AD process research, in particular, is the least explored area. AIM: We will address the two research gaps described above by conducting an empirical study of the audio description scripting process focusing on how it might be affected by experience of describers. URGENCY: This initiative is not only highly relevant to complement existing approaches in AD research, but it is also urgent. The development of process-oriented studies in AVT&MA is a crucial consideration to avoid a one sided view on AVT&MA practices as a product and elicit essential questions that product oriented approaches cannot answer (such as the motivation behind translation choices or the impact of contextual features on translation choices, to name but a few). The field of AD is currently at a pivotal point in its development, in need of solid, empirical evidence to underpin its theoretical claims and evolve into a mature discipline and the development of empirical process research is a crucial step in this development. Against this background, this project is not only a fundamental and logical next step in developing new knowledge in the field, but it will also contribute to establishing the University of Antwerp as a reference center in AVT and MA research. APPROACH: Our project is situated in an empirical paradigm and within the realm of Cognitive Translation and Inerpreting Studies. Building on methodological standards and measures of Translation Process Research this project is the first one to combine a comprehensive set of quantitative and qualitative tools to study the AD production process. INNOVATION AND IMPACT: (1) Research into a new, unclaimed research avenue within AVT&MA. (2) A multimethod experimental-design, linking methodologies rarely or never used within AVT&MA. (3) Interdisciplinary approach establishing collaboration within the University of Antwerp (TriCS, Antwerp Social Lab and Antwerp Human Lab). (4) tackling of the replication and reproducibility crisis through open science approach.

Researcher(s)

Research team(s)

Project type(s)

  • Research Project

Interacting minds, interacting bodies: Research infrastructure for psychophysiological sensor technologies and applications. 01/06/2022 - 31/05/2026

Abstract

This project is geared towards discovering and developing new applications of state-of-the-art psychophysiological sensor technologies (using computational and AI techniques) to help people with different needs work, learn and play in our modern society, ensuring that this tracking is meaningfully and responsibly applied. To accomplish this, our consortium is suitably interdisciplinary. This undertaking requires well-controlled lab studies and (near-)continuous psychophysiological tracking in real-life settings 'in the wild'. The research infrastructure applied for enables flexible movement from lab explorations of promising markers to checking their robustness in realistic, ecological contexts, and back again.

Researcher(s)

Research team(s)

Project type(s)

  • Research Project

Translation and interpreting in the global, digital age: Designing quality standards in audio description 01/10/2020 - 30/09/2025

Abstract

While audio description (AD) becomes one of the most widely researched AVT modalities, AD quality does not seem to have attracted the interest of many researchers so far. This stands in stark contrast to research on other modalities, especially subtitling and live subtitling, where quality is a pivotal issue and has been driving the sharp increase of audience reception studies in AVT (e.g., Di Giovanni & Gambier, 2018). AD quality is not merely an underrepresented topic in university research, but also a cogent problem in industry. The overreaching aim of the project is to create clear evaluation procedures of AD, based on first extensive and long-term analysis of AD quality. To do that we will attempt to define AD quality and map the relationship between AD scripting process and AD quality. While the project aims at introducing new themes in AD research (quality, AD process research) it also proposes employing research methodology that so far has been rarely used in AD research (keystroke logging, heart rate recording, presence questionnaires, interview, focus group, think aloud protocols, screen and face recording).

Researcher(s)

Research team(s)

Project type(s)

  • Research Project

Interlingual respeaking in the face of changes and new trends in interpreting and accessibility. 01/06/2024 - 30/11/2024

Abstract

Background: Live subtitling is a common method used to minimize sensory and language barriers. Initially, live subtitles were only intralingual (e.g. Dutch speech > Dutch subtitles) and catered for persons with hearing impairment. However, currently, there is a growing demand for interlingual live subtitles (e.g. Polish speech > English subtitles), which, in addition to hearing impaired persons, would also cater for all those that need access to foreign language. Aim: The aim of the project is to investigate the potential of interlingual live subtitles as an alternative to interpreting and discover how this method can be implemented to meet the expectations of different audiences. The research hypothesis is that, in some contexts, interlingual subtitling through respeaking is a feasible alternative to conference interpreting. Urgency: Social and technological changes may limit traditional translation services. Interlingual live subtitles have the potential to replace traditional interpreting, particularly in online and TV contexts, benefiting people with hearing disabilities. It also responds to modern challenges (globalization, digitalization, automation, aging of the population, wars, migrations, wider use of social media by gen Z) and complements simultaneous interpreting amid increasing automation. Approach: The project adopts a descriptive and comparative approach with four main research objectives: mapping interlingual live subtitling practices, creating a profile of a Polish interlingual live subtitler, investigating the quality of interlingual live subtitles compared to simultaneous interpreting, and analyzing audience reception. The author's professional involvement in live subtitling provides a unique perspective of a practisearcher. Innovation and impact: The project addresses current research gaps, introduces innovative methodologies, and highlights the societal impact of interlingual live subtitles on promoting inclusivity and equal access to information. It identifies future research directions, including exploring audience engagement and comprehension beyond other established quality measurements. By incorporating psychophysiological measures and leveraging emerging technologies, future research can enhance the effectiveness of interlingual live subtitles, ensuring broader industry involvement and innovation.

Researcher(s)

Research team(s)

Project type(s)

  • Research Project

Self-perceptions of Audio Describers: Does How You See Yourself Affect How You Describe? 01/11/2022 - 31/10/2024

Abstract

Audio description (AD) is a media accessibility service for non-sighted viewers, in which visual information is converted into an audio commentary. While this task is considered a form of intersemiotic translation (translating images into words), no knowledge of a foreign language is required to perform the task. In most countries audio describers come from a variety of backgrounds, not necessarily translation related. Therefore, the question arises: how do AD practitioners conceptualise what they do? This project looks to both translation and writing process research to gain insight into the process of drafting an AD script and the impact of the describer's self-perceived professional identity on both the process and the final product. This will be the first project of its kind, as process research in AD is extremely limited, despite calls for such studies (Jankowska 2021; Ramos and Rojo 2020; Braun 2007). Understanding the underlying processes of AD scripting will create new knowledge for both translation process research and writing studies and can be applied to design future AD training courses. Moreover, no previous study has considered the self-identity of AD practitioners. While attempts have been made to understand the skills and competencies of describers (ADLAB PRO 2017), this project will add a new, socio-psychological dimension to our understanding of the profession.

Researcher(s)

Research team(s)

Project type(s)

  • Research Project

Towards an inclusive Covid-19 crisis communication policy in Belgium: the development and validation of strategies for multilingual and media accessible crisis communication 01/02/2021 - 31/03/2022

Abstract

The interdisciplinary project subsidized by Sciensano aims to develop an effective strategy for more inclusive (digital) crisis communication, which takes account of the socio-linguistic diversity of Belgium and actively battles information inequality. The focus lies on how government communication about Covid-19 information during the pandemic can be improved through bespoke (re)translations and accessible media and language tailored to linguistic minorities' needs and specific needs groups. These needs include the information's linguistic/multimodal form, the communicative channels and dissemination measures.

Researcher(s)

Research team(s)

Project type(s)

  • Research Project

Designing quality standards in audio description. 01/10/2020 - 30/09/2024

Abstract

While audio description (AD) becomes one of the most widely researched AVT modalities, AD quality does not seem to have attracted the interest of many researchers so far. This stands in stark contrast to research on other modalities, especially subtitling and live subtitling, where quality is a pivotal issue and has been driving the sharp increase of audience reception studies in AVT (e.g., Di Giovanni & Gambier, 2018). AD quality is not merely an underrepresented topic in university research, but also a cogent problem in industry. The overreaching aim of the project is to create clear evaluation procedures of AD, based on first extensive and long-term analysis of AD quality. To do that we will attempt to define AD quality and map the relationship between AD scripting process and AD quality. While the project aims at introducing new themes in AD research (quality, AD process research) it also proposes employing research methodology that so far has been rarely used in AD research (keystroke logging, heart rate recording, presence questionnaires, interview, focus group, think aloud protocols, screen and face recording).

Researcher(s)

Research team(s)

Project type(s)

  • Research Project