Co-Designing Systems to Support Blind and Low Vision Audio Description Writers

“It’s Kind of Context Dependent”: Understanding Blind and Low Vision People’s Video Accessibility Preferences Across Viewing Scenarios

“…• What do you think just happened in the video? -If participants had simple questions or misconceptions, we clarified by concisely providing visual information, based on guidance in prior AD work [58].…”

“…Additionally, through evaluations of an automated AD system, Wang et al [126] found that BLV users preferred different details for different video types -for example, they wished to have more detailed descriptions of people in a comedy but not in a DIY video. Furthermore, Jiang et al [58] found that BLV AD writers focused specifically on character descriptions (e.g., race, age), background settings, action descriptions, and clarifying audio cues, given their experience as both creators and consumers.…”

“…• Level of Detail: Prior work has investigated the potential for visual question answering systems to enable users to query for details that they wish to know [6,58,110]. As AI advances, it may one day be possible to provide end users with high degrees of flexibility for which details and what level of detail they would like through automatically generated descriptions.…”

Making Short-Form Videos Accessible with Hierarchical Video Summaries

“…Videos can also be made more accessible using alternative modalities such as haptics or sound. Prior work has demonstrated the benefit of using haptics to conveying spatial information (e.g., location of actors on screen and facial expressions) in movies [71] and 360-degree videos [37]. Foley, or the reproduction of everyday sound effects added to videos to enhance audio quality, can artificially add rich information to the audio track of videos [13].…”

“…Web Content Accessibility Guidelines (WCAG 2.0) recommend that to make videos accessible, authors can create a summary of the content or add audio descriptions that synchronously narrate the visual content while avoiding overlapping with important audio content [72]. To support authors creating audio descriptions, prior work has proposed manual [36,40], collaborative [51], and (semi-)automated [15,17,24,46,52,60,73,76] approaches to create descriptions for a range of videos including long-form traditional films and TV shows [17,24], user-generated videos [36,46,51,52,60,73], livestreams [38,40], and 360-degree videos [18,23,37]. Short-form videos often include continual audio such that clear gaps do not exist for adding audio descriptions.…”