🧠 AI Accessibility Assistant

💡 Inspiration

The modern web is a gateway to education, communication, and opportunity—but for millions of people with disabilities, it's still filled with barriers.

We were inspired by a simple but powerful idea:
What if the web could adapt to you—your voice, your needs, and your unique way of interacting with information?

That question became the driving force behind the AI Accessibility Assistant.

We looked at the struggles faced by individuals with:

Visual impairments
Dyslexia and other reading difficulties
Motor disabilities
Cognitive limitations or neurodivergence

Existing accessibility tools are often fragmented, expensive, or lack user-friendly interfaces.
We wanted to build an all-in-one, AI-powered browser extension that combines the power of voice, vision, and personalization to make the web truly inclusive.

🛠️ How We Built It

We developed AI Accessibility Assistant as a Chrome extension using the following tools and technologies:

🔧 Tech Stack

JavaScript / HTML / CSS – Core frontend and browser extension logic
Chrome Extension APIs – For interacting with tabs, storage, and permissions
Web Speech API – For speech recognition and voice commands
Chrome TTS API – For converting text to speech
OpenDyslexic Font Integration – For dyslexia-friendly text rendering
ARIA Attributes & Contrast Enhancer – For improved visibility and navigation

🧠 Core Features Implemented

Feature	Description
🗣️ Text-to-Speech	Reads aloud text when user hovers or selects content.
🔤 Dyslexia-Friendly Font	Replaces fonts with OpenDyslexic to support readability.
🌗 High Contrast Mode	Applies contrast-optimized themes for visual impairments.
🎙️ Voice Commands	Allows users to navigate the web using voice: scroll, read, stop, etc.

🚧 Challenges We Faced

Like any ambitious project, we ran into several obstacles:

Voice Recognition Accuracy
Different accents and background noise posed challenges for consistent command detection.
We fine-tuned command parsing and added fallback mechanisms to improve robustness.
Cross-site Content Access
Chrome security policies restricted TTS implementation on arbitrary pages.
We used activeTab permissions wisely and leveraged content scripts.
Font Replacement Conflicts
Many sites use deeply embedded styles. Injecting OpenDyslexic without breaking layout
required precise DOM targeting and CSS specificity.
Real-Time UI Responsiveness
Enabling multiple features could reduce performance. We optimized our code for smooth
experience, even on low-end systems.

📚 What We Learned

This project was not just about coding—it was about empathy, inclusion, and human-centric design.

Key takeaways:

The power of universal design—building for one group often improves experiences for all.
How accessibility standards (like WCAG) can be enforced programmatically.
That AI doesn't need to be complex—small assistive features can have a big impact.

🚀 Looking Ahead

We’re excited about the potential of this tool and plan to:

Add sign language video overlays
Integrate screen magnifiers and braille output support
Launch on Firefox and Edge stores for broader reach
Use machine learning to personalize accessibility settings based on user behavior

🙌 Final Thought

“Accessibility is not a feature. It's a right.”

With the AI Accessibility Assistant, we’re taking one step closer to a barrier-free digital world—
where everyone can browse, learn, and thrive, regardless of their abilities.