Sentence Frames: Semi-Automating Alt Text without AI

Date & Time

Friday, March 22, 2024 - 10:20 AM PDT  

Location

Grand AB  

Description

AI-powered image description is a hot topic in accessibility. Advances are being made. But machines aren’t yet capable of producing contextual, audience-centered descriptions for figures such as graphs and charts.

Luckily, there are non-AI methods for improving the efficiency of large-scale figure description. In this session, Todd Marquis Boutin explores one such method: sentence frames.

A “sentence frame” is a fill-in-the-blank sentence traditionally used in academic assessments and language learning. However, for images such as xy-graphs and bar charts, sentence frames can be combined to create a customizable “description template” in Excel. The human operator enters relevant data in the input fields of the description template. Formulas then merge the input data and sentence frames into a cohesive description.

Inputting data is easier and faster than manually authoring a complete description, especially for complex figures. But sentence frames offer other benefits as well.

• More consistent language, style, and structure
• Reduced likelihood of introduced errors
• Large-scale pre-ingestion language editing
• Compatibility with other efficiency-improving techniques such as symbol and phrase substitution
• Possibility of user-customized figure descriptions

Through examples, Todd demonstrates the use of sentence frames and description templates for image types commonly found in electronic textbooks, journals, and online courses.

This Presentation Link is provided by the Presenter(s) and not hosted by the Center on Disabilities at CSUN. The Center on Disabilities has confirmed, as of April 18, 2024, content linked is relevant to the presentation, but has not been reviewed for accessibility nor will the Center on Disabilities attempt to remediate any accessibility issues in the linked content. Please contact the Presenter(s) with any accessibility concerns.

 

Audience

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

Audience Level

Intermediate  

Session Summary (Abstract)

Can AI create contextual descriptions of graphs and charts? Not yet. But we can semi-automate human-authored descriptions by using sentence frames. Learn how to build and use Excel-based description templates that convert input data into alt text.  

Primary Topic

Education  

Secondary Topics

• Blind/Low Vision
• Captions & Transcription
• Digital Accessibility
• Information & Communications Technology (ICT)

 

Session Type

General Track  

Presenter

• Todd Marquis Boutin
  Straive

Add to Outlook/iCal (How it works)

Back to Session List