Decoding AI for Accessibility

Lizzie Eastham and Sam Rickard present Studio 1 - Vision Australia Radio’s weekly look through a different lens.

On this week’s show

“Decoding AI for Accessibility”

Sam and Lizzie go all-in on AI – for better or worse.

It’s is an acronym that has gone from being the stuff of science fiction to mild confusion; moderate fear and face it... has ended up being all of the above and also very useful.

So we asked AI to come up with a title for the episode (Sam prefers his “Let me be your Brains”) and if you want to know more read the attached article... surprise surprise, it also has ben generated by AI!

Studio 1 welcomes any input from our listeners. If you have any experience or thoughts about issues covered in this episode or believe there is something we should be talking about.

EMAIL: studio1@visionaustralia.org or leave comment on the station’s Facebook page: https://www.facebook.com/VARadioNetwork

A special thank you to Stephanie; Carolyn; Jennifer and Garth

Next week: if you have a disability, there can be challenges to finding and maintaining friendships. We decided to get some professional Help. If you want to have your say on this topic or anything else please join our Whatsapp group here: https://chat.whatsapp.com/FwBORTly2OqC9FFmQAm97z?s=hd&p=i&mlu=3&amv=2

This program was made possible with support from the Community Broadcasting Foundation. Find out more at https://cbf.org.au/

Decoding AI for Accessibility: How Vision Australia Radio’s Studio 1 Explores the New Frontier of Assistive Tech

In a recent episode of Studio 1 on Vision Australia Radio, hosts Sam Rickard and Lizzy Eastham dive deep into artificial intelligence, exploring how an acronym once associated with science fiction has quickly transformed into an everyday lifesaver for people who are blind or living with low vision. Rather than treating AI as a distant novelty, the episode highlights practical applications, community stories, and the nuanced line between genuine empowerment and over-reliance.

Everyday Visual Description and Navigation

For many blind and low-vision listeners, visual description tools powered by AI have revolutionized daily social interaction and independence. Show contributors share how visual AI models give detailed context to images sent by family and friends over messaging apps. Rather than missing out on shared visual moments, users can upload photos to tools like Be My Eyes or ChatGPT to receive immediate, granular scene descriptions—from family photos to intricate environmental details like a full moon over a jetty.

Beyond standard static image descriptions, AI tools are expanding into real-time environmental navigation and daily tasks. App features such as Seeing AI and Aira allow users to quickly read mail, identify product labels, and recognize colors. Assistive apps linked with smart wearables like Meta glasses, such as the Orion app, offer continuous pathway updates—notifying users of obstacles like rubbish bins or helping them locate mislaid items like AirPods on the floor. Hands-free voice-driven tools like Lumen also streamline email management for Gmail users without requiring screen interaction.

Writing, Speechwriting, and Custom Software Creation

AI has also taken on a significant role in creative and professional workflows. Guest speakers detail how large language models like ChatGPT and Claude assist in structuring emails, formatting reports, and refining Toastmasters speeches. Rather than replacing human thought, contributors emphasize using AI as an editing partner to polish tone, organize interview questions, or brainstorm creative speech concepts.

Taking technological empowerment a step further, computer instructor Garth Humphries demonstrates how AI allows non-coders to build bespoke software tailored to their visual accessibility needs. Using Claude, Humphries developed a customized audio-video editing suite and custom controls for a Stream Deck controller to solve specific accessibility barriers that off-the-shelf software failed to address.

The Pitfalls: Audio Editing, AI Slop, and Loss of Creativity

Despite the overwhelming benefits, Studio 1 doesn't shy away from the downsides of emerging AI technology. Co-host Sam Rickard shares his experience testing AI editing software like Descript, pointing out that automated filler-word removal often cuts off natural speech patterns, creating unnatural edits. The hosts also discuss Adobe Podcast's speech enhancement tools, demonstrating how AI noise reduction can strip away spatial ambiance if over-applied.

Furthermore, the episode touches on broader cultural concerns surrounding "AI slop"—monotonous, machine-generated content flooding platforms like YouTube and Spotify. The hosts advocate for maintaining authentic human storytelling in literature, art, and music, reminding listeners that while AI serves as an exceptional assistant, human perspective remains irreplaceable.

See omnystudio.com/listener for privacy information.