Logo of the Unlimited conference

Unlimited! 3 - Innovation for Access: New Interactions

Unlimited! 3 Innovation for Access: New Interactions took place online on February 11, 2022.

The Unlimited! conferences are an opportunity to discuss research, practice, policy and technological innovation regarding the accessibility of live events and broadcasts. Stakeholders from varying backgrounds and domains come together to pinpoint challenges, share best practices and explore convergences and new collaborations in a growing range of services and products that make live events and broadcasts more accessible for diverse audiences - think of audio description, audio surtitling, audio subtitling, surtitling and sign language. The first Unlimited! conference was organized by the TricS research group of the University of Antwerp on 26 April 2016. The second Unlimited! conference was held in Barcelona on 6 June 2018. It was organized by the Autonomous University of Barcelona (UAB) and the EU-funded project Accessible Culture and Training (ACT), in which the University of Antwerp was a partner. In 2021, we also organised a free online teaser event in anticipation of our face to face event. In 2022, Unlimited! 3 took place online, and was  preceded by the OPEN Forum on 10 February, an exhibition and networking event around the theme of Media Accessibility.
 
The provision of accessibility services for live events is gaining prominence throughout the world. It is a vibrant and dynamic domain, replete with innovative approaches. Unlimited 3! contributed to the new opportunities created by these innovative trends, by exploring recent developments in access services such as audio description, audio surtitling, audio subtitling, surtitling for the hearing impaired, and sign language. Two key seem to be driving the field forward: technology and creativity. Technological innovations have always been in the provision of pre-recorded access services, but lately technology’s role is growing in live access as well. On the other hand, new creative approaches to live access are developing rapidly; approaches that look for new ways to integrate accessibility into creation processes and that focus on engagement and on creating experiences for all audiences. However, these new opportunities also pose major challenges. Stakeholders have to learn to navigate the variations and complexities that characterize this approach. These complexities create new opportunities, but also challenge all parties involved, requiring them to move out of their comfort zone, go beyond their individual expertise and foster new, interdisciplinary collaboration. All these developments are giving rise to new types of practices, new types of collaboration, new professional profiles, and new requirements for training. The third Unlimited! symposium aimed to explore these challenges and innovative solutions in live accessibility and bring together stakeholders to share experiences. 
 
A post conference survey distributed among participants, showed that the overall quality, organization and accessibility rate for the conference was considered good to excellent. Survey participants unanimously answered "yes" when asked if they would encourage others to attend next time. Despite the conference being held online through Zoom, survey participants commented positively on the networking opportunities. One of the conferences’ main aims was to connect a wide variety of stakeholders to exchange ideas. In this respect, the participation of artists in the conference was appreciated by survey participants, and the OPEN team hopes such dialogue across profiles and disciplines can be strengthened in future events.
 

THE Unlimited! 2022 speakers

Keynote 1 by Joselia Neves: MOVES AND SHIFTS TO MITIGATE MISMATCHES… with due respect and humble submission.

Session 1: Audio Description I

  • Implementing Audio Description in Video Games, by Xiaochun Zhang.
  • Craftsmanship and technology: two sides of the story in audio description, by Irene Hermosa.
  • Describing personal characteristics in audio introductions – towards equitable lexical choices, by Alina Secara and Raluca Chereji.

Session 2: Subtitling/captioning

  • Speech-to-text interpreting (STTI) in Sweden: Research and Practice, by Ulf Norberg.
  • When accessibility meets multimedia learning: Effect of intralingual live subtitling on perception, performance and cognitive load in an EMI university lecture, by Yanou Vangauwbergen, Isabelle Robert & Iris Schrijver.
  • Evaluating Novel Systems for Improved Captioning of Non-Speech Audio in Video Content, by Lloyd May.

Keynote 2 by Hannah Thompson: Audio Description as ‘Blindness Gain’: Beyond Access towards Co-Creation

Session 3: Accessibility

  • The IDE@ project: inclusive and accessible online teaching, by Estella Oncins.
  • Evaluating automatic spoken language translation as a tool for multilingual inclusion, by Claudio Fantinuoli.
  • Artificial Voices for Audiovisual Translation and Media Accessibility – Present and Future Scenarios, by Alexander Kurch.

Session 4: Audio Description II

  • Audio Description in Kung Fu Scenes: A Case Study of An Auteur Description Production in Hong Kong, by Luo Kangte and Jackie Xiu Yan.
  • How do I look? Reflections on the challenges and opportunities of audio describing a fashion show, by Polly Goodwin.
  • The Audio Describer As Cast Member: Audio Description At Every Performance, by Joel Snyder.

Closing session

  • From the Universal to the Self in Media Accessibility: Accessibility as a Promise, by Pablo Romero Fresco & Kate Dangerfield.

Read the full programme with abstracts below!

 

Sponsors & supporting organisations

The OPEN Forum was made possible with the support of the Department of Applied Linguistics, Translators and Interpreters and the Council of service provisiosn (Raad Dienstverlening) of the University of Antwerp and with the support of our accessibility advisors and sponsors:

Gold Sponsors

The Audio Description Company, a division of The Subtitling Company.

Logo of the Audio Description Company

 

Earcatch, a free app that offers audiodescriptions for movies and TV-series.

logo van earcatch

 

Ooona offers a range of ground-breaking cloud based translation software.

Logo van Ooona

 

Silver Sponsor

Limecraft, workflows for video production.

Logo van Limecraft in witte letters op een limoengroene achtergrond. Daaronder de slogan connected creativity.

 

Supporting organizations

Conférence Internationale Permanente d'Instituts Universitaires de Traducteurs et Interprètes (CIUTI)

Logo of CIUTI

The department of Applied Linguistics, Translators and Interpreters and the Council for service provision (Raad Diensverlening) of the University of Antwerp.

 

Accessibility Advisors

Hoorcoach Regina Bijl

Logo Hoorcoach Regina Bijl

Flemish expertise centre Inter .

Logo Inter

 

OPENING KEYNOTE: MOVES AND SHIFTS TO MITIGATE MISMATCHES… with due respect and humble submission, by Joselia Neves (Hamad Bin Khalifa University of Qatar)

Every time we mediate communication, in whatever form it takes, our goal will be to mitigate possible or existing mismatches between the source text and its receiver. It is believed that translation has performed the task with great ability. The desire to ‘assist’ somebody who ‘lacks’ the ability to access the original message has also, for long, characterized the work of those working in mainstream accessibility. Recent years have seen a turn towards creative approaches that place mediation beyond the functional purpose of serving the receiver. At times, the liberties taken have led to the creation of new originals, initially thought to respond to the needs of a specific target group, but that have deviated from the source text to the point of being seen as a totally different artifact, with an identity of their own. In parallel, other accessibility services position themselves within the realm of technical translation, leaving very little space for creativity, constraining the mediator to find strategies that will respond to the receiver’s needs in the given context. In this presentation we will look at two instances in which audio description has been taken to the extremes of information and expression to see how they may be two sides of the same coin. We will use the achievement space framework (Neves 2018) to infer the moves and shifts that occur in terms of the parameters of delivery and engagement, to evaluate to whom (or what) we, the translators/mediators, ‘pay due respect and humble submission’.
 

SESSION 1: AUDIO DESCRIPTION I

The Audio Describer As Cast Member: Audio Description At Every Performance, by Joel Snyder (Audio Description Associates, LLC-Audio Description Project of the American Council of the Blind – U.S.)
In 1981, a formal audio description service—the world’s first—was begun under the leadership of a blind woman at The Metropolitan Washington Ear, a radio-reading service based in Washington, DC. Radio reading services still exist throughout the United States with the participation of volunteer readers; I began working as a volunteer reader at The Ear in 1972 and was proud to be a founding member of its audio description service. Radio reading services are heavily dependent on volunteers and The Ear’s audio description service was also structured around voluntary contributions of time and effort. Cognizant of the limits on the times of people who often maintain full-time employment elsewhere, audio description was conceived as a service that would be offered at only two performances of a theatrical run and preparation for the audio described performances was based on the observation of only two or three performances early in the run. It was understood that optimally audio description would be prepared with more in-depth observation of the theatrical event, even during rehearsals, and that audio described should be offered at every performance in the run of a show. But the limitation of the volunteer structure prohibited that arrangement. The proliferation of audio description for live theatrical events in the United States is based on this volunteer, limited preview/two-described-performances model. With support from the D.C. Aid Association for the Blind, the Audio Description Project of the American Council of the Blind proposed a more expansive audio description arrangement for two productions again at Arena Stage. We collaborated with Arena on the experiment: an audio describer attended rehearsals for each production, met with the stage director, actors, the designers (scenic, costumes, lighting, sound) and developed an audio description script throughout the three-week rehearsal period. The script was then available for that same describer to voice at every performance beginning with opening night and with, of course, an eye on stage action as minor changes in action could occur from performance to performance. The describer, essentially, was a “cast member”, attending every rehearsal and performance. The arrangement had two benefits over the traditional model of audio description development for live performance: time was available to carefully observe the theatrical process and construct descriptive language that was more thorough and considered; and people desiring the service could attend any performance with no advance notice and be assured of access to the visual aspects of the production. Other innovations included Braille and large-print programs, models of the set and props in the lobby, and a tactile “scrapbook” of costume pieces. It was gratifying to note that attendance for the productions by people using audio description tripled over levels experienced at Arena using the traditional volunteer model.
 

Axel, Elisabeth and Nina Levent (2003). Art Beyond Sight: A Resource Guide to Art, Creativity, and Visual Impairment. New York: Arts Education for the Blind and AFB Press

Charlson, Kim and Berk, Judy (2001). Making Theater Accessible: A Guide to Audio Description in the Performing Arts, Northeastern University Press, June 2001

Clark, Joe, "Standard Techniques in Audio Description": unpublished but accessible at http://joeclark.org/access/description/ad-principles.html

Greening, Joan and Deborah Rolph (2007). “Accessibility - Raising awareness of audio description in the UK”. Díaz Cintas, Jorge Pilar Orero & Aline Remael (eds) Media for All: Subtitling for the Deaf, Audio Description and Sign Language. Amsterdam: Rodopi

Holland, Andrew (2009). “Audio Description in the Theatre and the Visual Arts: Images into Words”. In Gunilla Anderman and Jorge Díaz-Cintas Audiovisual Translation. Language Transfer on Screen. Basigtoke: Palgrave Macmillan

Snyder, Joel (2008). “The Visual Made Verbal”. In Jorge Díaz Cintas (Ed) The didactics of Audiovisual Translation. Amsterdam: John Benjamins

York, Greg (2007). “Verdi Made Visible. Audio-introduction for Opera and Ballet”. In Jorge Díaz Cintas, Pilar Orero & Aline Remael (eds.). Media for All. Accessibility in Audiovisual Translation. Amsterdam: Rodopi

 
Craftsmanship and technology: two sides of the story in audio description, by Irene Hermosa (Università Autònoma de Barcelona, Spain)
This paper aims to present and reconcile two current approaches audio description (AD) aimed at visually impaired users in the context of Spain. On the one hand, technological advancements such as automatically cued and text-to-speech AD (Fernández-Torné & Matamala, 2015; Walczak & Fryer 2018; Hermosa-Ramírez, 2020) are increasingly being applied not only to assistive technologies such as e-readers and screen readers, but also to the scenic arts and films. On the other hand, “artisan” AD involving creators (Cavallo, 2015; Fryer, 2018) and users (see Di Giovanni, 2018, for more on participatory accessibility) are increasingly being advocated for. Based on the results of a recent online focus group within the RAD research project (Researching Audio Description: Translation, Delivery and New Scenarios), the argument put forward is that there is room for both approaches, but they ought to be applied for different products or purposes. Previous research has already demonstrated that text-to-speech AD is deemed more acceptable for certain audiovisual genres (Szarkowska, 2011; Walczak & Fryer, 2018). In the scope of a recent focus group with end users of an older age, participants put forward a clear distinction: they were satisfied with text-to-speech solutions for everyday use, i.e., assistive mobile applications, but strongly rejected its mainstreaming in products and cultural activities such as audio books, audio described films and theatre shows. Second, a case study is presented to show how technology and “craftmanship” can co-exist even in the same activity or production. In August 2019, Barcelona’s city council offered an array of access services for Gràcia’s district festivities. Amongst them, an audio described guided tour of the adorned streets of Gràcia was on offer. As an alternative, visually impaired visitors could use the ‘Wheris’ app to explore the streets autonomously whenever they chose. By examining this proposal, we also advocate for the inclusion of technical advances that allow for a repeatable, pre-recorded approach for static productions. This paper ultimately aims to argue that a quantitative increase in AD via new technologies should not be the one and only goal. It needs to be in parallel with more creative, more participatory approaches that prioritize quality. Users are requesting it, we as researchers and practitioners should follow suit.
 

Barcelona Accesible: http://ajuntament.barcelona.cat/accessible/es/tematicas/fiestas-y-celebraciones.

Cavallo, A. (2015). Seeing the word, hearing the image: the artistic possibilities of audio description in theatrical performance. Research in Drama Education: The Journal of Applied Theatre and Performance, 20(1), 125–134.

Di Giovanni, E. (2018). Participatory accessibility: Creating audio description with blind and non-blind children. Journal of Audiovisual Translation, 1(1), 155–169.

Fernández-Torné, A. & Matamala, A. (2015). Text-to-speech versus human voiced audio description: A reception study in films dubbed into Catalan. The Journal of Specialised Translation, 24, 61–88.

Fryer, L. (2018). The independent audio describer is dead: Long live audio description! Journal of Audiovisual Translation, 1(1), 170–186.

Hermosa-Ramírez, Irene (2020). Delivery approaches in audio description for the scenic arts. Parallèles, 32(2), 17–31.

Szarkowska, A. (2011). Text-to-speech audio description: Towards wider availability of AD. The Journal of Specialised Translation, 15, 142–162.

Walczak, A. & Fryer, L. (2018). Vocal delivery of audio description by genre: Measuring users’ presence. Perspectives: Studies in Translatology, 26(1), 69–83.

Describing personal characteristics in audio introductions – towards equitable lexical choices, by Alina Secara and Raluca Chereji (University of Vienna, Austria)
VocalEyes, a leading organization providing primarily live audio descriptions for theatrical plays in the UK, recently published the Describing Diversity Report in partnership with Royal Holloway (Hutchinson, Thompson, and Cock 2020). Through questionnaires, interviews with professional audio describers and theatre creatives, and using a corpus of 26 audio introductions (AIs), it explored when and how personal characteristics of characters that appear on stage are described. In particular, the report focused on the visible, physical markers of race, gender, disability, age and body shape. Starting from the premise that language choices reflect certain beliefs and biases, their aim was to look for ways to create future equitable and inclusive audio description. It is in this context that we set to test some of the report’s findings and enrich it with further data. We analyze a corpus of 53 English AIs produced by five UK professional audio describers which cover a variety of genres including comedies, musicals, pantomimes, and drama. Using corpus investigation tools integrating part-of-speech tagging and lemmatization, we provide an analysis of the linguistic features of the personal characteristics present in our corpus. We offer an investigation into potential bias and imbalances in the rendering of personal characteristics, focusing in particular on elements of implicit and explicit gender, age and race. In particular, we look into the trends identified in the VocalEyes Report, such as explicit or implicit mentioning of Whiteness and Blackness, the use of gendered adjectives to describe body shapes and abilities, as well as the presence of value judgements in relation to gender and age and compare them against our corpus. In addition, we also contribute to investigations into the structure of AIs. Previous research suggests a possible structure for AIs for theatre (York 2007) and calls for standardization of AI structure in opera (Iturregui-Gallardo and Solás 2019). We combine findings from two previous studies (Di Giovanni 2014; Reviers, Roofthooft, and Remael 2021) which identified general event information, general performance information, production-specific information, synopsis, scenography and characters as main AI elements, to investigate their presence and consistency of use in our corpus. Our contribution, therefore, is twofold: one the one hand, we supplement with linguistically richer investigations some of the findings within the Describing Diversity Report regarding the description of personal characteristics related to gender, age, and race in AIs. On the other hand, we provide empirical evidence related to the structural and chronological consistency of the AI content elements identified in previous studies. Our findings contribute to both research and training in audio description. 
 

Di Giovanni, Elena. 2014. ‘Audio Introduction Meets Audio Description: An Italian Experiment’. InTRAlinea, no. Special Issue: Across Screens Across Boundaries.

Hutchinson, Rachel, Hannah Thompson, and Matthew Cock. 2020. ‘Describing Diversity Report. An Exploration of the Description of Human Characteristics and Appearance within the Practice of Theatre Audio Description’. VocalEyes.co.uk/about/research/describing-diversity.

Iturregui-Gallardo, Gonzalo, and Iris Cristina Permuy Hércules de Solás. 2019. ‘A Template for the Audio Introduction of Operas: A Proposal’. Hikma, no. 18: 217–35.

Reviers, Nina, Hanne Roofthooft, and Aline Remael. 2021. ‘Translating Multisemiotic Texts: The Case of Audio Introductions for the Performing Arts’. JosTrans. The Journal of Specialised Translation, no. 35.

York, Greg. 2007. ‘Verdi Made Visible: Audio Introduction for Opera and Ballet’. In Media for All: Subtitling for the Deaf, Audio Description, and Sign Language, edited by Jorge Díaz Cintas, Pilar Orero, and Aline Remael, 215–30. Amsterdam: Rodopi.
 
SESSION 2: SUBTITLING/CAPTIONING
 
Speech-to-text interpreting (STTI) in Sweden: Research and Practice, by Ulf Norberg (Stockholm University, Sweden)
Sweden has been organizing vocational training courses for skrivtolkar, i.e. speech-to-text-interpreters, since the 1970s. Since the 1980s d/Deaf and hard-of-hearing persons have been entitled to access STTI services in a wide range of settings, including healthcare, education and training, contacts with social services, government agencies, in legal contexts, work-related meetings, for cultural events, religious ceremonies and leisure time activities. While live subtitling for broadcasts and live events, and in particular the use of respeaking, have attracted significant academic attention in recent years, STTI in other settings is still largely unexplored. To inspire research into STTI and develop a blueprint for best practice, we have produced a handbook of Speech-to-Text Interpreting, which brings together contributions from STTI practitioners, educators, researchers and users. The handbook provides an in-depth insight into the practical work of STT interpreters at both single-speaker events and conversational exchanges, discussing, amongst others, the use of different keyboards and technologies, strategies for preparation and collaboration, techniques for faster typing speeds, remote interpreting, work procedures and types of equipment to current and future developments in STTI. It also provides an overview of international research into STTI, which is underpinned by a translation/interpretation studies perspective, and examines issues of orality and literacy as well as ethical aspects. While the handbook is primarily intended for students, teachers, practitioners and users of STTI, it is also a guide for event organizers, educational institutions, government agencies and others interested in the improved provision of accessibility. In this presentation, we give an overview of the evolution and content of the Handbook and present suggestions for further research.
 
When accessibility meets multimedia learning: Effect of intralingual live subtitling on perception, performance and cognitive load in an EMI university lecture, by Yanou Van Gauwbergen, Isabelle Robert & Iris Schrijver (University of Antwerp, Belgium)
One of the main challenges in higher education in the 21st century is providing educational access to an increasingly multilingual and multicultural student population. Many universities are therefore considering using English as language of instruction (EMI), but students’ limited proficiency in English can be a serious drawback. Live subtitling might help to overcome this language barrier, by removing physical (auditory) and linguistic barriers at the same time. The aim of this paper presentation is to report on preliminary results of a project that investigates (1) how university students in Flanders perceive EMI lectures with intralingual live subtitles, i.e. lectures for which the words of the lecturer are subtitled in real time in the same language (English), (2) whether these subtitles influence their performance, and (3) what impact these subtitles have on their cognitive load. In this project, the impact of subtitling on perception, performance and cognitive load have been investigated during five two-hour Marketing lectures taught in English to students of Economics who have Dutch as their mother tongue. The live subtitling was produced in real time through respeaking (using speech recognition software) each time during two lecture fragments of approximately 25 minutes (one before the break and one after the break in each lecture). Quantitative and qualitative data have been collected using (1) online language tests, consisting of a certified listening test and vocabulary test to determine the students’ English proficiency; (2) online questionnaires on demographics (e.g., mother tongue and self-reported proficiency in English); (3) comprehension tests after each lecture about the content (and the perception) of the lecture; (4) eye tracking glasses to measure cognitive load; (5) post-hoc interviews after the lecture series. Data are now being analyzed. In this presentation, we will mainly report on the quantitative data.

 

Garcia, G. D. (2021). Data visualization and analysis in second language research. Routledge.

Lasagabaster, D., & Doiz, A. (2021). Language use in English-medium instruction at university: International perspectives on teacher practice. Routledge.

Mayer, R. E. (2021). Multimedia learning: Third edition. Cambridge University Press.

Robert, I. S., De Meulder, A., & Schrijver, I. (2021). Live subtitling for access to education: A pilot study of university students’ reception of intralingual live subtitles. JoSTrans, 36a, 53–78.

Xiaoming, X., & Norris, J. M. (2021). Assessing academic English for higher education admissions. Routledge

 
Evaluating Novel Systems for Improved Captioning of Non-Speech Audio in Video Content, by Lloyd May (Stanford University, U.S.)
Closed captioning in video content for non-speech audio, such as music or sound effects, often has a frustratingly low level of detail. Sound artist Christine Sun Kim illustrates these limitations as well as the creative possibilities of captioning in her work as an artist and curator [1-2]. Inspired by her work, we asked if other techniques could provide more pertinent information to the viewer than words alone overlayed on-screen. Previous research has explored both screen-based solutions including stand-alone visualizations for music [3-4], screen overlays in video games [5-6], and projection and lighting around the screen [7]. However, the generalized case of a screen overlay style visualizer for video content has not been studied extensively. We conducted need-finding interviews with Deaf and Hard of Hearing (DHoH) participants to understand the strengths and limitations of current captioning for non-speech audio. Participants highlighted inconsistencies across viewing platforms, lack of temporal information such as beat and song endings, and lack of clarity in emotion conveyed by music as issues of particular concern. Additional genre-specific issues were raised, such as tension conveyed via music in horror and intentionally juxtaposed music in comedy. The results from these initial user interviews were used to generate a list of attributes of non-speech sounds that users expressed interest in having captioned: volume, beat, melody, directionality, mood, and title of the piece of music. In consultation with two Deaf technologists, we developed several screen overlay prototypes. The prototypes consist of a graphic overlay on the sides of the screen with visual elements that change in real-time in response to properties of the non-speech audio. The system allows for captioning artists to map audio properties to visual attributes of the overlay. The audio properties include computed properties such as beat onset, loudness, and spectral roughness, and captioning artist-defined properties including perceived emotion, valence, and function within the video content. The mappable visual attributes include color, line weight, movement, smoothness, and density. We then surveyed users with a range of self-identified hearing ability statuses, with an emphasis on recruiting DHoH individuals to evaluate the screen overlay prototypes as compared to traditional captioning, expressive/poetic captioning, and sign-language interpretation. Participants viewed short clips from a range of genres and were asked to rate the performance of the captioning/interpreting technology according to its legibility, information density, and how distracting it was. Additionally, participants rated the relative importance of each attribute of non-speech sound in each clip to generate a preferred information hierarchy. An individualized, configurable system is necessary for an experience that contains information that a user finds sufficient, but not overly distracting. While this system of tailored audio-reactive visual overlays has the capacity for a degree of automation, it is designed to allow for manual programming that utilizes the artistry and creativity inherent in a captioner’s practice.

 

Sun Kim, Christine. “Artist Christine Sun Kim Rewrites Closed Captions | Pop-up Magazine.” YouTube, 13 Oct. 2020, https://www.youtube.com/watch?v=tfe479qL8hg.

Van Tomme, Niels, and Christine Sun Kim. “Activating Captions - ARGOS.” Activating Captions | ARGOS Centre for Audiovisual Arts, 4 June 2021, https://www.argosarts.org/activatingcaptions/info.

Deja, Jordan Aiko, et al. "ViTune: A Visualizer Tool to Allow the Deaf and Hard of Hearing to See Music With Their Eyes." Extended Abstracts of the 2020 CHI Conference on Human Factors in Computing Systems. 2020.

Ho-Ching, F. Wai-ling, Jennifer Mankoff, and James A. Landay. "Can you see what I hear? The design and evaluation of a peripheral sound display for the deaf." Proceedings of the SIGCHI conference on Human factors in computing systems. 2003.

Holloway, Alexandra, et al. "Visualizing audio in a first-person shooter with directional sound display." 1st Workshop on Game Accessibility: Xtreme Interaction Design (GAXID'11) at Foundations of Digital Games, Bordeaux, France. 2011.

Collins, Karen, and Peter J. Taillon. "Visualized sound effect icons for improved multimedia accessibility: A pilot study." Entertainment Computing 3.1 (2012): 11-17.

Jones, Brett R., et al. "IllumiRoom: peripheral projected illusions for interactive experiences." Proceedings of the SIGCHI Conference on Human Factors in Computing Systems. 2013.

 

AFTERNOON KEYNOTE by Hannah Thompson

Audio Description as ‘Blindness Gain’: Beyond Access towards Co-Creation
In 2021, UK learning disabled theatre group Mind the Gap created an audio-described version of their film a little space. Rather than bringing in an external provider to add an audio description track to the finished film, the company created the description themselves, using a ‘speaking out loud’ technique they had developed during rehearsals. Also in 2021, a group of blind, partially blind and non-blind people co-wrote a creative audio description of an artwork in the musée du quai Branly’s ‘Gularri: waterscapes from Northern Australia’ exhibition. Together, these two co-created audio descriptions raise important ethical questions around autonomy, authority and representation which will be explored in this paper. Using the concept of ‘blindness gain’, we will discuss the benefits and disadvantages of inclusive, co-created description with reference to several examples from the IDEA project. We will explore what happens when audio description moves from access provision to artistic intervention and ask who has the right to describe or be described. We will consider the potential of novel techniques such as overt positionality, multiple descriptions and the audio describer’s ‘preface’ before thinking about what is at stake when audio description is repositioned as an essential part of the artistic experience for both blind and non-blind beholders.
 
The IDE@ project inclusive and accessible online teaching, by Estella Oncins (Università Autònoma de Barcelona, Spain)
The COVID-19 crisis has impacted education at all levels, with online distance education becoming central due to the forced transition from face-to-face to online environments. Yet, it is essential to go beyond the emergency remote teaching practices, distinguish between the highly variable online teaching solutions (Hodges et al. 2020), and develop sustainable, inclusive and accessible online practices. In the last decades, there has been a growing and accompanying awareness of the need to promote social inclusion, and authorities are setting specific requirements and legislation for teaching institutions to improve inclusion and accessibility of the teaching and learning materials and methods. The United Nation Convention on the Rights of Persons with Disabilities (UN CRPD, 2006) in its article 24 “recognize the right of persons with disabilities to education. With a view to realizing this right without discrimination and on the basis of equal opportunity”. This is a major challenge for many educational systems worldwide, especially in online teaching environments, in which there is a need for a comprehensive view of the pedagogy of online education that integrates technology to support teaching and learning (Carrillo & Assunção-Flores, 2020). In this sense, a widely used framework is the Universal Design for Learning (UDL), a framework of teaching and learning planning that helps teachers to create lessons that are inclusive for a broad range of learners in their classrooms (CAST, 2014). Digital accessibility and inclusive learning are the central axes of the IDE@ project, which is an Erasmus+ project (2021-2023) that aims at developing the necessary skills to train professionals in online synchronous, asynchronous, and hybrid educational contexts, to create inclusive and accessible online teaching materials to reach all students from a UDL perspective. The project is divided into four phases. The first phase aims to identify the challenges of distance learning in terms of accessibility from a learners' perspective. The second phase aims to identify the challenges of distance learning in terms of accessibility from the perspective of training professionals. In the third phase, unified guides will be generated for the creation of a new professional profile "Expert in the development of inclusive and accessible materials for distance training". The fourth phase will certify the competencies of this new professional profile for both the academic and vocational fields. This will ensure the quality and sustainability of the project at European level. The project is currently in the first and second phases in which two different and complementary data collection techniques are planned. First, surveys on current synchronic, asynchronic, and blended online distance contexts will be carried out for the detection of main challenges associated to accessible and inclusive teaching and learning practices. Second, focus groups with training professionals and learners will be conducted for the validation and enrichment of data obtained from the online surveys to create a certified modular curriculum. 

Carrillo, C. & Assunção-Flores, M. (2020). “COVID-19 and teacher education: a literature review of online teaching and learning practices”. European Journal of Teacher Education, 43:4, 466-487, DOI: 10.1080/02619768.2020.1821184

Center for Applied Special Technology (CAST). (2014). What is universal design for learning? Wakefield, MA: Center for Applied Special Technology. Retrieved from http://www.udlcenter.org/aboutudl/whatisudl

Hodges, C., S. Moore, B. Lockee, T. Trust, & Bond, A. (2020). “The Difference between Emergency Remote Teaching and Online Learning.” EDUCAUSE Review. Retrieved from https://er.educause.edu/articles/2020/3/the-difference-between-emergency-remote-teaching-and-online-learning

United Nations. (2006). The convention on the rights of persons with disabilities and its optional protocol (A/RES/61/106). Retrieved from https://www.un.org/development/desa/disabilities/convention-on-the-rights-of-persons-with-disabilities.html

Artificial Voices for Audiovisual Translation and Media Accessibility – Present and Future Scenarios, by Alexander Kurch (Freelance audiovisual translator)
Creating extensive services for accessible audiovisual contents still face financial obstacles. This especially concerns productions that require studio recordings with professional speakers. An exemplary case is audio description for the blind and visually impaired. But there is a cost-effective option that could potentially allow more productions like these: Artificial voices - also known as speech synthesis or text-to-speech technology (TTS for short). Due to significant quality improvements in recent years, the range of application scenarios for speech synthesis has already grown beyond the media industry. The increasing everyday use and acceptance of speech technologies show that it is worthwhile to look at current and future use cases. In the context of audiovisual translation (AVT), TTS as a production option for auditive contents is of interest for audio description, audio subtitles, voice-overs, dubbing and game localization. However, generic out-of-the-box solutions often cannot meet the qualitative expectations for some of these media products. As in the evolving application of automatic speech recognition and machine translation for intralingual and interlingual subtitling, a use case-specific and targeted implementation of these technologies with human post-editing and quality assurance is needed. Therefore, before using artificial voices for audiovisual contents, a range of questions arise regarding the successful integration of this technology: For which audiovisual formats is speech synthesis of reasonable use? For which scenarios has TTS technology for accessible media already been applied? How can the quality of artificial voices be measured or assessed? Can artificial voices be improved by some form of post-editing? While some of these aspects will be covered in my presentation, additional issues need to be discussed for future practice, like: How can artificial voices be integrated with the original audiovisual image and sound material? Which tools and competences are necessary for working with TTS? Can speech synthesis be used for (semi-) live events? Another user-centred and thus decisive question is, for which products does this alternative auditory reproduction of information not only find application, but acceptance? As an example, my presentation briefly outlines the use case of audio description with speech synthesis and the role of TTS post-editing.
 
Evaluating automatic spoken language translation as a tool for multilingual inclusion, by Claudio Fantinuoli (Mainz University/KUDO, Germany)
Language accessibility has become increasingly important over the last few years. Public and private stakeholders are called to take action and increase the inclusion of both people with hearing impairment, learning disabilities, and cognitive restrictions as well as of people that do not share the language of the community they live in. In the context of multilingual communication, human-crafted accessibility services, such as interlingual subtitling (Romero-Fresco and Pöchhacker, 2017) and interlingual interpreting (Pöchhacker, 2016) have emerged as instruments of inclusion. Since those high-quality services are available only for a limited number of events, live automatic spoken language translation (SLT) could have the potential to further diminish communicative barriers and increase social inclusion. Recent advances in artificial intelligence have led to dramatic improvements in the quality of automatic SLT (cf. Bojar et al. 2021; Sperber et al. 2020). Little is known about how such systems may perform in real communication settings, for example, if they can be used without losing functional adequacy and how their approaches to user-machine interaction may influence the perception by the end-users (cf. Karakanta et al. 2021). Most prominently, there is a lack of evaluation frameworks that take into consideration all these communicative aspects. Against this backdrop, we present an improved communicative-oriented framework to evaluate automatic speech-to-text translation based on the one proposed by Fantinuoli and Prandi (2021). The framework aims at assessing the quality of SLT on the basis of a comparison to human simultaneous interpretation, considered as the ‘communicative’ gold standard in a real-world scenario of multilingual communication. We illustrate the results of an empirical experiment to evaluate real-time SLT using this framework and how it compares with automatic metrics used in the field. We assess its potential and limitations as a tool for the evaluation of SLT. Prospective developments in the proposed framework are also illustrated. 
 

Bojar, Ondřej, Vojtěch Srdečný, Rishu Kumar, Otakar Smrž, Felix Schneider, Barry Haddow, Phil Williams, and Chiara Canton. “Operating a Complex SLT System with Speakers and Human Interpreters.” In Proceedings of the 1st Workshop on Automatic Spoken Language Translation in Real-World Settings (ASLTRW), 23–34. Virtual: Association for Machine Translation in the Americas, 2021. https://aclanthology.org/2021.mtsummit-asltrw.3.

Fantinuoli, Claudio, and Bianca Prandi. “Towards the Evaluation of Automatic Simultaneous Speech Translation from a Communicative Perspective.” In Proceedings of the 18th International Conference on Spoken Language Translation (IWSLT 2021), 245–54. Bangkok, Thailand (online): Association for Computational Linguistics, 2021. https://doi.org/10.18653/v1/2021.iwslt-1.29.

Karakanta, Alina, Sara Papi, Matteo Negri, and Marco Turchi. “Simultaneous Speech Translation for Live Subtitling: From Delay to Display.” In Proceedings of the 1st Workshop on Automatic Spoken Language Translation in Real-World Settings (ASLTRW), 35–48. Virtual: Association for Machine Translation in the Americas, 2021. https://aclanthology.org/2021.mtsummit-asltrw.4.

Pöchhacker, F. (2016). Introducing Interpreting Studies. Manchester: Routledge.

Romero-Fresco, P., & Pöchhacker, F. (2017). Quality assessment in interlingual live subtitling: The NTR Model. Linguistica Antverpiensia, 16, 149–167.

Sperber, Matthias, and Matthias Paulik. “Speech Translation and the End-to-End Promise: Taking Stock of Where We Are.” ArXiv:2004.06358 [Cs], April 14, 2020. http://arxiv.org/abs/2004.06358.

 
SESSION 4: AUDIO DESCRIPTION II
How do I look? Reflections on the challenges and opportunities of audio describing a fashion show, by Polly Goodwin (Freelance (PolySensoryAccess) and Vision Australia)
Fashion has a critical role to play in our lives – spanning the prosaic need to keep warm and decent through to a medium of high art, all the while acting as an expression of who we are (be that through deliberate careful curation of an outfit right or a necessary reflection of our socio-economic circumstances). Yet in many respects, fashion (and the ultimate medium for its presentation and celebration: the fashion show) remains one of the last bastions of inaccessibility for people who are blind or have low vision. There have been a mere handful of attempts to give meaning to fashion shows for participants who are blind or have low vision, effectively excluding this cohort from engaging with a presentation that represents art and zeitgeist. Audio Description is the art and craft of translating the visual into the spoken word. Aimed, in the first instance, at consumers who are blind or have low vision, it has the potential to enable this group to enjoy and authentically experience an event alongside their sighted peers – be that theatre, ballet, visual art, film, television or any other presentation containing a significant visual component. From March 2020, Vision Australia has partnered with the Melbourne Fashion Festival to provide audio description for key runways and, as the describer tasked with planning and executing the experience, in this paper I will be charting my journey in bringing together the contexts of fashion and blindness and low vision aimed at delivering a truly meaningful experience for all concerned. Examining the theoretical and practical opportunities and challenges (including those thrown up by the COVID19 pandemic, such as cancellation of live runways at 2 hours’ notice and a shift to online), and reflecting the needs and feedback of participants, the aim is to start the process of providing some guidelines and recommendations for other describers to build on to bring the fashion show firmly into the audio description stable.
 
Audio Description in Kung Fu Scenes: A Case Study of An Auteur Description Production in Hong Kong, by Luo Kangte & Yan Jackie Xiu (City University of Hong Kong)
By translating images into words, audio description (AD) provides opportunities for people with visual impairments to access media products. In traditional AD production, an audio describer is usually a third-party person that doesn’t belong to the filmmaking team (Fryer, 2018a). In recent years, the role of film directors in the AD creation process has received increasing attention (e.g., Udo & Fels, 2009; Szarkowska, 2013; Fryer, 2018b; Romero-Fresco, 2020). The term “auteur description” is used to refer to a type of AD that incorporates the director’s creative vision in the script (Szarkowska, 2013) and gives audio describer artistic license to depart from the dictate of objectivism. However, there is a dearth of empirical studies on existing auteur description products. In recent years, local NGOs in Hong Kong have cooperated with local filmmaking films and jointly produced more than 30 DVDs. Some ADs are created under the supervision of film directors (Tsui, 2019). For example, the AD script of the Chinese martial arts film Yip Man: The Final Fight (Yau, 2013) was jointly drafted by the screenwriter and the director, and the screenwriter voiced the AD lines. The description of the kung fu fighting scene was highly appreciated by the visually impaired audience in Hong Kong. The current study conducts textual analysis on the AD of the kung fu fighting scene of Yip Man: The Final Fight. By transcribing and analyzing the 3-minute audio described clip, the researchers identified five distinctive features that differentiate it from traditional AD : 1) using vivid and varied voice that matches the plot development, 2) using evaluative adjectives, 3) using simile that enhances the narrative coherence, 4) describing the inner thoughts of the characters, and 5) using jargon and terminology. After illustrating the five features with examples, the research will discuss how these findings would influence AD teaching in a university in Hong Kong. For example, apart from being guided to describe what is shown on screen, students should be led to search for contextual information of the film (e.g., through the screenplay, interviews, reviews) and integrate the information into AD scriptwriting. This study is expected to shed light on the development of AD production and training in Hong Kong and other areas.

 

Fryer, L. (2018a). The independent audio describer is dead: Long live audio description! Journal of Audiovisiual Translation, 1(1), 170-186.

Fryer, L. (2018b). Staging the audio describer: An exploration of integrated audio description. Disability Studies Quarterly, 38(3).

Romero-Fresco, P. (2019). Accessible filmmaking: Integrating translation and accessibility into the filmmaking process. London and New York: Routledge.

Szarkowska, A. (2013). Auteur description: From the director's creative vision to audio description. Journal of Visual Impairment & Blindness, 107(5), 383-387.

Tsui, Y. S. (2019). The voice linking the two worlds – the ten years of audio description in Hong Kong. Hong Kong: Joint Publishing (H.K.).

Udo, J. P., & Fels, D. I. (2009). “Suit the action to the word, the word to the action”: An unconventional approach to describing Shakespeare's Hamlet. Journal of Visual Impairment & Blindness, 103(3), 178-183.

Yau, H. (Director). (2013). Yip man: The final fight [Film]. National Arts Films Production; Emperor Film Production.

 
Audio Description in Video Games, by Xiaochun Zhang (University of Bristol, U.K.)
Video gaming has evolved into one of the world's most favorite forms of entertainment over the past few decades. There are 2.7 billion active gamers in the world who generate $174.9 billion in revenue for the gaming industry in 2020. However, most video games are not accessible or fully accessible for people with disabilities, who account for 15% of the world population (WHO, 2018). Game accessibility for players with sight loss is especially challenging due to the visual and interactive nature of games. Making games accessible for blind and visually players requires all visual elements to be represented by means of auditory or haptic feedback. Audio description can be one of the solutions to improve game accessibility (Mangiron & Zhang, 2016). So far, AD has been applied in video game trailers and live game streaming, where gamers/streamers record themselves playing video games to a live audience online. This presentation introduces the AD4Games project, which investigates how AD can improve game accessibility by working with professional audio describers, game developers, and visually impaired participants. We use the game Before I Forget, developed by two of our partners, to conduct a few experiments, testing ways AD can enhance game accessibility. The AD4Games project first investigates how AD can be applied to game streaming by producing and seeking feedback on three types of AD: 1) recorded AD of a game playing video; 2) live AD of a game streaming session; 3) live AD of an audio describer playing the game, acting both as audio describer and game streamer. Secondly, the project explores the feasibility of AD-assisted game playing. Visually impaired participants are invited to play the game with audio describers who describe visual elements in the game live based on the players’ actions. Based on feedback and data collected from interviews, surveys, focus groups, workshops, and observation, the team will co-adapt the game into a more accessible version and test it with visually impaired participants. This presentation will report the research outcomes of this project and discuss the pending issues in game accessibility.
 

Mangiron, C. & Zhang, X. (2016). Game Accessibility for the blind: Current overview and the potential application of audio description as the way forward. In A. Matamala & P. Orero (Ed.). Researching audio description: New approaches, pp. 75-95. London: Palgrave Macmillan.

WHO. (2018). Disability and health. https://www.who.int/news-room/fact-sheets/detail/disability-and-health

CLOSING SESSION

Accessible Theatre Performances – A Semiotic Approach, by Maria Wünsche (University of Hildesheim, Germany)
In the last two decades, theatre accessibility has not only been gaining prominence in practice, but also in research. Studies on this topic stem from Cultural Studies (Ugarte Chacón 2015), Disability or Deaf Studies (Vollhaber 2018), as well as from AVT (Griesel 2014, Mälzer/Wünsche 2018). In AVT, a focus has long been on access tools as add-ons or, following Greco (2019), “ex-post solutions” that are developed after the production process. Integrating accessibility proactively in the production process, however, is gaining considerable momentum: In cultural practice, this approach is called Aesthetics of Access (Sealey/Lynch 2012). It refers to access tools being used not only in a pragmatic or linguistic, but also in a creative and artistic way. This changes not only the production and collaboration process, but might also alter traditional expectations towards theatre captioners, audio describers, and other access facilitators. So far, AVT has come up with different methods to tackle this issue: In the case of captioning, the terms creative (McClarty 2012) or creactive subtitling (Robert 2016) have been introduced. The concept of co-translation (Mälzer 2016) refers to the necessary collective production process. Another approach worth investigating is the semiotic analysis of the artefacts and translation processes involved in such theatre productions. In the ImPArt project by the Un-Label Performing Arts Company, an adaptation of Antoine de Saint Exupéry’s “The Little Prince” (cf. https://un-label.eu/en/project/the-little-prince/) was developed following the Aesthetics of Access approach. This has already been presented shortly at the 2021 Unlimited! 3 Teaser Event. The play was developed in an English-speaking work environment, but for an audience in Athens, Greece. It was created as a multisensory performance with integrated captioning, vibrating benches, and olfactory elements. From a Translation Studies perspective, it is especially interesting to analyze because its production entailed various intersemiotic, intralingual and interlingual (Jakobson 1971, Agnetta 2019) translation processes, e.g. from the original literary text to the theatre script and from the theatre script to English and later Greek captions or AD elements. These tasks were furthermore carried out by different stakeholders involved in the project. The presentation will focus not only on these processes and stakeholders, but also shed light on the interplay of different semiotic modes on stage during the live event and their impact on translation processes. A
 

Agnetta, Marco (2019): Ästhetische Polysemiotizität und Translation. Glucks Orfeo ed Euridice (1762) im italienisch-deutsch-französischen Kulturtransfer. Hildesheim: Georg Olms.

Gian Maria Greco (2019): Accessibility Studies: Abuses, Misuses and the Method of Poietic Design. In: HCI (LBP), 15-27.
Griesel, Yvonne (2014): Welttheater verstehen. Berlin: Frank & Timme

Jakobson, Roman (1971): Selected writings. Volume II: Word and Language. Berlin, New York: De Gruyter

Mälzer, Nathalie (2016): Audiodeskription im Museum. Ein inklusiver Audioguide für Sehende und Blinde. In: Barrierefreie Kommunikation – Perspektiven aus Theorie und Praxis. Berlin: Frank & Timme, 209-229.

Mälzer, Nathalie/Wünsche, Maria (2018): Inklusion am Theater. Übertitel zwischen Ästhetik und Translation. Bern: Peter Lang.
McClarty, Rebecca (2012): Towards a multidisciplinary approach in creative subtitling. In: MonTI (4), 133–153.

Robert, Èlia Sala (2016): Creactive subtitles: subtitling for ALL.
Sealey, J./Lynch, C. H. (2012): Graeae: An Aesthetics of Access – (De)Cluttering the Clutter. In: Broadhurst/Machon: Identity, Performance, and Technology. Practices of Empowerment and Technicity. Basingstoke: Palgrave Macmillan.

Ugarte Chacón, Rafael (2015): Theater und Taubheit. Ästhetiken des Zugangs in der Inszenierungskunst. Bielefeld: transcript.

From the Universal to the Self in Media Accessibility: Accessibility as a Promise, by Pablo Romero Fresco & Kate Dangerfield (University of Roehampton, London, U.K.)
Film, in terms of practice and theory within the industry and academia, has a long-standing relationship with the notion of universality. In the industry, the notion that cinema is a universal language has been used on “both a domestic and international level [...] as a tool of control and colonization” to dominate the scene for economic purposes (Dwyer, 2005: 304). The concept of universal also denies difference and marginalizes or excludes subjects/fields (people), which further complicate its premise (Dwyer, 2005; Eleftheriotis 2010). Audiovisual translation, media accessibility and more specifically the accessible filmmaking model undermine the notion of universality in film, as they illustrate difference by rendering translation and translators visible and highlighting the inaccessibility of cinema both as a practice and as an institution (Romero-Fresco, 2019). However, through the reference to universal design and the use of the “for all” tag used in legislation, conferences and publications over the past years, media accessibility and accessible filmmaking have somehow reinstated the notion of universality, which may sometimes mask the exclusion of certain users who are not catered for by most mainstream guidelines. This is the case for the people with dual/single sensory impairments and complex communication needs involved as co-researchers and filmmakers in the research and documentary carried out by one of the authors of this proposal (Dangerfield, 2020), which requires the reconsideration of what is meant by difference and brings the notion of the universal further into question. The first aim of this presentation is to explore how this notion of universality has been used in media accessibility and the impact it has had on research, training and practice in this area. Particular attention will be paid to the currently prevailing cognitive and experimental turn experienced by audiovisual translation and media accessibility over the past decade (Chaume, 2018, Greco et al., 2020). This has been critical to provide a solid scientific basis for widely used guidelines that until now had been based on the personal views of experts. While these studies are as necessary now as they have been over the past ten years, it is also worth looking at what has fallen by the wayside. The emphasis on statistical significance means that we have preferred to learn “a little bit about a lot of participants” than “a lot about a few participants”. We have been more concerned with what is shared by all (hence the use of the “for all” tag) than with what sets us apart. This has been useful to consider participants as a group, but it has left behind some individuals, particularly those who do not conform to the general patterns and who are discarded as outliers. This has resulted in guidelines that, while valuable and democratic, encourage a homogeneous use of MA, regardless of the individual nature of every film and of the idiosyncrasy of the viewers. Now that these guidelines are consolidated and regularly included in training courses, it may be time to explore what lies beyond them (for instance, more creative approaches to media accessibility) and how to allow for the participation of those who are still left behind. The second aim of this presentation is to introduce and analyze the work of an emerging wave of (mostly disabled) artists who are proposing an alternative approach to media accessibility, one that is openly subjective, increasingly creative and that often works as a political tool in a wider fight against discrimination and for real inclusion (Romero-Fresco, 2020 and forthcoming). Special emphasis will be placed on the work produced for live events and museums by artists such as Christine Sun Kim, Liza Sylvestre, Caroline Lazard and Chisato Minamimura. Many of these alternative and creative practices question the objective and static nature of access, presenting it as a promise and as speculative practice, rather than as a guarantee (Lazard, 2019). The presentation will conclude with a series of pointers suggesting how this approach can be applied in media access training and research and with a proposal to use the notion of “media for all” as an aspiration or end goal accompanied by a series of questions that mitigate the risks inherent to most claims to universality in film, translation and access.

 

Contact

Steering committee

Nina Reviers, Sabien Hanoulle, Aline Remael, Isabelle Robert, Gert Vercauteren, Anna Jankowska.

 

Scientific committee

Pablo Romero-Fresco (University of Vigo)
Agnieszka Szarkowska (University of Warsaw)
Louise Fryer (University College London; VocalEyes)
Anna Matamala (Autonomous University of Barcelona)
Pilar Orero (Autonomous University of Barcelona)
Elena Di Giovanni (University of Macerata)
Elisa Perego (University of Trieste)
Rai Sonali (Royal National Institute of the Blind)
Nathalie Mälzer (University of Hildesheim)
David Mass (Panthea)
Andreas Tai (Institut für Rundfunktechnik München)
Maaike Bleeker (Universiteit Utrecht)
Anna Jankowska (Autonomous University of Barcelona)
Hayley Dawson (University of Roehampton)
Jemina Napier (Catholic University of Leuven)
Heidi Salaets (Catholic University of Leuven)
Gert Vercauteren (University of Antwerp)

 

You can contact us at open@uantwerpen.be with any questions you may have.