Enabling Real-Time Document Accessibility Using AI

Date & Time

Wednesday, March 20, 2024 - 3:20 PM PDT  

Location

Elite 1-3  

Description

Lack of timely access to visual information is a significant problem for the nearly 24 million blind or visually impaired (BVI) individuals in the U.S. and 285 million globally. Significant progress has been made in ensuring BVI individuals get the necessary accommodations they need. However, very little work has been done towards making critical materials accessible in real-time, especially for non-textual materials such as math expressions and graphical representations. Currently, authoring, converting, and/or producing accessible versions of non-textual materials requires Braille transcribers and tactile graphic designers with rare and specialized knowledge. This reliance on human-driven processes leads to high costs and significant delays from the time the accessible materials are needed to when they are actually delivered to BVI individuals. Artificial Intelligence (AI) systems have the potential to address this issue by automating the conversion/generation of accessible materials. To demonstrate the concept, we developed a novel AI-powered system, Kanak, that converts visual (scanned) math worksheets samples into accessible formats optimized for: (1) Embossers, (2) Screenreaders, (3) multisensory digital access on IOS mobile application, (4) Braille readers, and (5) audio devices. Initial evaluations with expert transcribers indicated significant time savings and efficiency in authoring accessible versions of math worksheets using this inclusive solution.  

Audience

• Higher Education
• Information & Communications Technology
• K-12 Education
• Research & Development

Audience Level

Intermediate  

Session Summary (Abstract)

BVI individuals have challenges accessing non-textual materials such as math and graphical representations in real-time. We demonstrate how AI systems can automate the process of conversion/generation of accessible materials in real-time.  

Primary Topic

Artificial Intelligence (AI) & Machine Learning (ML)  

Secondary Topics

• Blind/Low Vision
• Digital Accessibility
• Education

 

Session Type

General Track  

Presenters

• Hari Palani
  UNAR Labs
• Nicholas Giudice
  The University of Maine

Add to Outlook/iCal (How it works)

Back to Session List