Annotation as a Tool for Accessibility for Blind and Vision Impaired Students
Thanks to Assistive Technologies based on Text-To-Speech (TTS) and refreshable braille displays, textual content on the Open Web Platform and in EPUB 3 textbooks can be accessible to students who are blind. However, despite HTML5 and WAI-ARIA supporting various mechanisms for image accessibility, publishers typically fail to add alternative textual descriptions or don't include sufficiently comprehensive descriptions, particularly in educational materials. Graphical charts have also typically lacked the underlying data that blind students can consume. And mathematical content has been particularly challenging as only recently have publishers begun to author using MathML. Thus a large body of legacy mathematical content remains as inaccessible images unless cost-effective solutions can be found. This content could be annotated as MathML and correspondingly described using rules sets, such as MathSpeak.
Benetech sees annotations as a technology that can be used by special education teachers, Disabled Student Services offices, parents and other members of the students support network to add descriptions or transcriptions to knowledge encapsulated in images as a post-production, remediation step. For this use case, annotations will need:
- Support for annotating bitmap and SVG images (In the case of SVG, being able to annotate one or a group of elements would be ideal.)
- Accessible annotation reading, navigation, and authoring tools.
- Mechanisms for screen readers and other assistive technologies to inform the user that one or more annotations providing alternative modalities are available
- Mechanisms for blind and vision impaired users to request annotations that provide alternative modalities or representations of the visual content
- Support for HTML markup to describe complex images such as pie charts with tables.
- Support for MathML to transcribe images that are mathematical formulas (MathML is supported by various Assistive Technologies.)
- Metadata to identify the descriptions as alternatives or transcriptions of inaccessible or poorly described visual content.
- A mechanism for original publishers to query, analyze and integrate "crowdsourced" descriptions and transcriptions created by annotation in order to pull those back into the original content.
In order to provide students with print disabilities equal access to the general education curriculum, the DIAGRAM Center, a Benetech Global Literacy initiative, is conducting research and developing tools to improve the way image and graphic content for accessible instructional materials (AIM) is produced and accessed. The center is chartered to conduct research on the current state of accessibility of images and other non-text content in AIM; develop innovative, open-source technologies and tools to address the challenges of producing and displaying accessible content; and recommend practices and develop training to support these efforts.
Much of the work of the DIAGRAM Center and Benetech is relevant to the needs described by this use case. In this workshop we will:
- Demonstrate the DIAGRAM Center's Poet tool for crowdsourced description and transcription of visual content in DAISY and soon EPUB 3 eBooks.
- Discuss technologies for transcription of mathematical images that could be leveraged in annotation technologies.
- Describe tools that aid those inexperienced in image description to create useful image descriptions, particularly in educational materials.
We hope that workshop participants will take with them an understanding of this use case and corresponding technical requirements and engage in dialogue about how to best deliver on the technical requirements.