OmniAccess

Inspiration

Web browsing is essential, yet the default internet experience remains unoptimized for many:

1 in 5 people (over a billion individuals) live with dyslexia.
2.2+ billion people worldwide face visual impairments, a demographic expanding rapidly alongside an aging population. We were inspired to build OmniAccess to shift control directly to the user, creating an all-in-one Chrome extension that bridges the digital divide and transforms any webpage into an inclusive space for everyone.

What It Does

OmniAccess delivers an accessible, tailored browsing experience by combining instant visual customization with smart AI assistance directly from your browser toolbar.

Instant visual customization: Tweak font styles (including dyslexic font and bionic text), line spacing, color contrast, and cursor size in real time to meet individual needs.
Smart AI assistance: Dynamically summarize and simplify complex text to match the reader’s target comprehension level.
Automated image captioning: Instantly generate context and descriptions for image-heavy pages.
Customizable text-to-speech: Select specific blocks of webpage text to convert them into spoken audio with a single click.

How We Built It

We built OmniAccess as a manifest-based Chrome extension, using standard web technologies for the frontend interface and background scripts to handle real-time DOM manipulation.

Cognitive and visual AI features: Integrated the Gemma model from featherless.ai to do custom text modification, dynamically adjust reading levels, and generate live image captions.
Audio accessibility: Leveraged Web Speech API to deliver high-quality, natural text-to-speech rendering directly within the browser.
Project organization: Our team organized task distribution and workflow using Opennote, keeping our engineering process structured and ensuring efficient, structured execution.

Challenges We Ran Into

Designing an accessible interface for diverse users presented a steep learning curve for our team. We had to carefully research and implement specialized features, such as bionic reading, dyslexia-friendly fonts, custom spacing, and variable contrast, ensuring the layout remained intuitive across multiple demographics. On the technical side, we spent significant time debugging API integrations and solving CSS inconsistency issues across different web layouts. Overcoming these hurdles required tight time management and ultimately made us more thoughtful, resilient developers.

Accomplishments That We're Proud Of

We are incredibly proud to have built a comprehensive, multimodal accessibility tool that addresses digital barriers impacting nearly 25% of the global population. By combining visual and textual assistance powered by the Gemma model with realistic voice synthesis from Web Speech API, we seamlessly brought multiple cutting-edge AI features into a single browser extension. Successfully taking this project from concept to a functional tool that bridges the web accessibility gap is our team’s proudest achievement.

What We Learned

Gained hands-on experience integrating cutting-edge tools like text-to-speech engine workflows and real-time image captioning, technologies we are excited to build upon in future work. Mastered asynchronous team collaboration. Navigating a remote workflow required us to divide responsibilities strategically, conduct thorough code reviews, and maintain clear communication to keep our development process efficient and on-schedule. Deepened our understanding of accessibility standards by researching specialized reading modes, features, and inclusive design principles.

What's Next for OmniAccess

Looking forward, our goal is to launch OmniAccess on the Chrome Web Store to make it immediately available to everyday users. Beyond desktop browsing, we plan to expand into mobile app development, ensuring universal accessibility across all personal devices. As we grow, we will continue adding new assistive features and refining our AI capabilities to broaden our impact and reach even more users worldwide.