AccessEd

A web for everyone. Browser extension to assist the visually impaired.

Comment

Inspiration

AccessEd: A web for everyone We students from Algeria, Canada, and Indonesia(in Taiwan) have come together with the vision of building an accessible world for all. With majority of us living in countries with aging populations, we often see the struggles elderly experience accessing the internet. Observing how individuals with vision impairments were very common, even amongst youth, we decided to make the web more accessible to visually impaired individuals.

What it does: Our Main Features

We've built AccessEd, a browser extension that leverages AI, text-to-speech, and webpage content analysis to create a more accessible web.

Highlight offers instant outlines of contents hovered.

Bigger Cursor upgrades the cursor into a circular outline.

Read Aloud provides audio for text segment clicked on. Offers audio in various accents in English, as well as 3 types of audio speed.

Describe Image utilizes the Gemini 2.0 AI model to describe images hovered on. Providing context on image background, colors and objects.

How we built it

Highlight : We took specific interactive elements and inserted a border into their CSS file.

Front-end Development : With focus on accessibility our extension's settings page uses colorblind-friendly colors and large text, while our webpage uses React and is hosted with Vercel.

Back-end Development : Utilizes Flask for framework and PythonAnywhere for hosting.

AI Integration: Takes the image link and sends it to the back-end Flask server where image analysis with the Gemini API happens.

Challenges we ran into

Collaborating with a unified vision was our main challenge. As university students, we find it challenging to schedule meetings , especially with our very different time zones. With our lack of meetings, we had to individually develop features, our vision of this project developed separately from each other.

Describe Image : Due to a good amount of image links being restricted, some images cannot be analyzed by AI, in place their captions are provided.

Demo Webpage : With the challenges of scheduling meetings, our demo webpage might display features that have slight differences with the ones developed in our project.

By integrating various technologies, various functions and data became incompatible with each other, causing our project to develop a variety of bugs and errors. This forced us to remove features originally planned such as the feature Zoom Lens that enlarges text hovered on.

Accomplishments that we're proud of

We are thankful of having had each other throughout this project. Our achievements of AI integration with Gemini and successful debugging of various features have provided us a demo that displays our core ideas.

What we learned

Our project provided us with experience developing with various technologies and integrating them with each other, as well as challenged our problem solving skills. Although our collaboration is somewhat lacking, we learned how important brainstorming and feedback are.

What's next for AccessEd

We will continue in taking steps towards a more accessible and inclusive world. Firstly we will publish our extension to the Chrome Web Store.

Feel free to give feedback through our survey on Github, it is much appreciated. With it we will refine the current features and visuals, upgrading its user-friendliness and customization to specifically cater to the needs of individuals with visual impairments.

Built With

Share this project:

Updates