"It's Kind of Context Dependent": Understanding Blind and Low Vision People's Video Accessibility Preferences Across Viewing Scenarios (2403.10792v1)

Abstract: While audio description (AD) is the standard approach for making videos accessible to blind and low vision (BLV) people, existing AD guidelines do not consider BLV users' varied preferences across viewing scenarios. These scenarios range from how-to videos on YouTube, where users seek to learn new skills, to historical dramas on Netflix, where a user's goal is entertainment. Additionally, the increase in video watching on mobile devices provides an opportunity to integrate nonverbal output modalities (e.g., audio cues, tactile elements, and visual enhancements). Through a formative survey and 15 semi-structured interviews, we identified BLV people's video accessibility preferences across diverse scenarios. For example, participants valued action and equipment details for how-to videos, tactile graphics for learning scenarios, and 3D models for fantastical content. We define a six-dimensional video accessibility design space to guide future innovation and discuss how to move from "one-size-fits-all" paradigms to scenario-specific approaches.

• The paper reveals that BLV users’ video accessibility needs vary significantly with viewing context.
• It uses surveys and interviews to critique one-size-fits-all audio description methods across media types.
• The research introduces a six-dimensional design framework and discusses generative AI's role and limitations.

Understanding Video Accessibility Preferences for Blind and Low Vision Users

The paper “It's Kind of Context Dependent”: Understanding Blind and Low Vision People's Video Accessibility Preferences Across Viewing Scenarios” presents a comprehensive examination of how blind and low vision (BLV) individuals interact with and require modifications for video accessibility. This paper recognizes the limitations of traditional audio description (AD) approaches and advocates for more nuanced, scenario-specific methods to accommodate the diverse needs of BLV audiences across various video consumption contexts.

Overview of Research Objectives and Methodology

The research aims to elucidate BLV users' video accessibility preferences in diverse scenarios, addressing the inadequacies of the "one-size-fits-all" AD model traditionally used across media. The authors conducted a formative survey followed by semi-structured interviews with BLV participants to gather qualitative data on their experiences and preferences. The paper identifies significant gaps in the accessibility of varied video types such as educational, how-to, and entertainment content across different platforms like YouTube, Netflix, and social networking sites.

Key Findings and Scenario-Specific Preferences

The paper reveals that BLV users' accessibility needs and preferences vary greatly depending on the video type and context of viewing. For example, in educational settings, users prioritize detailed descriptions of visual aids and settings to facilitate comprehension. In contrast, entertainment such as music videos and dramas requires emphasis on characters, settings, and visual effects for an immersive experience. For rapidly evolving short-form content, synchronization of accessible content, such as extended descriptions or prologues, was particularly sought after.

Emerging Design Space for Video Accessibility

The researchers propose a novel six-dimensional design space for video accessibilities, such as:

1. Level of Detail: This ranges from minimal to extreme, depending on user preference for verbose descriptions.
2. Alteration of Video Time: Varied description durations can alter the pacing of video content for optimal user comprehension.
3. Level of Augmentation: This involves the degree to which videos are enhanced post-production with accessibility measures.
4. Modality of Presentation: Beyond spoken descriptions, modalities include visual enhancements, Braille, tactile models, and audio cues.
5. Synchronicity of Accessible Content: Timing of access features, potentially before, during, or after video consumption, is essential.
6. Tone and Style of Approach: This varies based on scenario, calling for narrative styles that align with user goals and content type.

Consideration of Generative AI in Video Accessibility

The paper astutely considers the role of generative AI technologies in expanding video accessibility options. By automating certain aspects of content description and enhancement, these technologies hold potential for broadening the scope of personalized accessibility accommodations. However, the authors caution against unregulated AI deployments due to potential biases and ethical considerations, underscoring the need for meticulous dataset curation and robust quality evaluation.

Implications for Future Developments

This research carries significant implications for the future of video accessibility. There is a compelling case for incorporating user-centered design into the creation of accessibility features, acknowledging not only BLV individuals' varied preferences across scenarios but also the rapid evolution of content platforms and viewing technologies. The paper highlights the necessity of adopting innovative approaches, such as integrating tactile and auditory feedback, to satisfy the diverse needs of BLV users in an increasingly digital landscape.

In conclusion, the paper provides a well-founded argument for a shift from uniform AD approaches to more flexible, context-aware video accessibility strategies. The proposed design space offers valuable frameworks for researchers and practitioners aiming to develop more inclusive and personalized media experiences for BLV audiences. As technology further interlaces with media consumption, these insights will become increasingly critical in guiding both practical applications and academic pursuits in human-centered computing and accessibility research.