Sounds Accessible: Envisioning Accessible Audio Media Futures with People with Aphasia

Research Background and Problem

• Identified Problems or Challenges: Audio media (e.g., radio and podcasts) play a crucial role in global events, education, and entertainment. However, for users with complex communication needs (e.g., individuals with aphasia), accessibility to audio media poses significant challenges. While research has focused on the accessibility of audiovisual media, studies specifically addressing audio media, particularly for users with complex communication needs, remain scarce.
• Importance of the Problem: Individuals with aphasia are affected by language impairments, which limit their ability to communicate and access information in daily life. Given the widespread use of audio media, this accessibility issue hinders their equal participation in information access and cultural experiences. Addressing this issue not only improves the quality of life for individuals with aphasia but also advances inclusive design practices that combat ability-based discrimination.
• Research Motivation and Related Work: The authors propose two primary questions: (1) How can interventions be designed to enhance the ability of individuals with aphasia to engage with and experience audio media? (2) How can these interventions be integrated into daily life? Additionally, the study explores the social dynamics of audio media consumption and how interventions might impact such experiences. This research fills a gap in studies on the accessibility of audio media for individuals with aphasia, offering new directions for the field of design.

Solution

• Proposed Methods or Solutions: The authors employed a "Research-through-Design" approach, conducting six co-design workshops and developing low-fidelity and high-fidelity technological tools to explore interactions between individuals with aphasia and audio media, ultimately designing novel accessibility solutions.
• Innovative Aspects of the Solution:
  1. Emphasizing the social and emotional roles of audio media beyond its functional use, including contributions to therapy and recovery.
  2. Proposing three major directions for improving audio media: multimodal enhancements (integrating visual or textual aids), content restructuring and simplification (e.g., summaries and segmentation), and designing entirely new formats tailored to the needs of individuals with aphasia.
  3. Introducing generative AI technologies (e.g., automatic summarization and image creation) to achieve personalization and accessibility.
• Implementation Steps and Key Technologies:
  1. Conducting participatory design workshops with individuals with aphasia to gather requirements and feedback.
  2. Introducing technological tools in later stages, such as "AccioBook" (speech rhythm control), "Simplico" (content simplification), and "IncluScribe" (multimodal visual aids).
  3. Leveraging generative AI technologies, such as OpenAI's GPT models, to generate text and images that enhance accessibility.
  4. Testing design prototypes with individuals with aphasia to iteratively refine the solutions.

Research Outcomes

• Specific Achievements:
  1. Gaining a deeper understanding of the barriers faced by individuals with aphasia in consuming audio media, such as cognitive load, language comprehension difficulties, and challenges with short-term memory.
  2. Establishing new design directions, including multimodal enhancements, content restructuring, and aphasia-friendly formats, along with specific design recommendations.
  3. Developing and testing technological probe tools, such as automatic rate control and content simplification features, to directly demonstrate innovative solutions for individuals with aphasia.
• Advantages Compared to Existing Solutions:
  1. Breaking away from traditional audio media's reliance on "normal speech fluency" by proposing user-centered innovative designs.
  2. Combining personalization and accessibility through generative AI, reducing reliance on traditional manual design.
  3. Incorporating users' emotional, social interaction, and recovery experiences into the design scope, expanding the boundaries of traditional accessibility research.
• Experimental or Evaluation Results:
  1. Workshops revealed the difficulties individuals with aphasia face in understanding complex language and lengthy audio content.
  2. Prototype tools, such as AccioBook's rhythm control and Simplico's speech simplification features, received positive feedback, particularly for reducing cognitive load.
  3. Visual aid tools (e.g., IncluScribe's image generation) were found to be helpful for comprehension, though image quality and accuracy require further improvement.
• Limitations and Future Directions:
  1. The long-term effects and real-world applications of the tools need further validation.
  2. Research is needed on the impact of errors in generative AI-generated content on users and methods for improvement.
  3. Exploring how these technological tools can be integrated into existing audio media platforms while preserving artistic integrity.
  4. Considering a broader range of user groups, such as other populations with complex communication needs.

Through this study, the authors not only propose a new framework for redefining audio media accessibility but also provide rich inspiration and pathways for future design and research.