AI-AccessibilityKit

AI-Powered Accessibility Toolkit for iOS Apps

AI-AccessibilityKit is a Swift package designed to make iOS apps more inclusive by leveraging Core ML, Vision, and SwiftUI. It automates accessibility audits and enhancements, helping developers deliver WCAG-compliant, VoiceOver-friendly, and Dynamic Type-ready experiences.

---

✅ Features

• Auto VoiceOver Labels
  Automatically generate accessibility labels for images and icons using:

  • Core ML image classification
  • Vision OCR for text detection

• Dynamic Font Scaling
  Apply adaptive font sizes for better readability across all Dynamic Type settings.

• Color Contrast Audit
  Real-time WCAG 2.1 contrast ratio checks with visual badges in previews.

• Touch Target Audit
  Highlights UI elements smaller than Apple’s recommended 44pt minimum.

• Real-Time Accessibility Overlay
  Adds a 44pt grid and audit badges in Xcode Previews for proactive design checks.

---

📦 Installation

Add AI-AccessibilityKit to your project using Swift Package Manager:

1. In Xcode, go to File → Add Packages…
2. Enter the repository URL:

https://github.com/Mulla6518/AI-AccessibilityKit.git

3. Select Add Package and link it to your app target.

---

🛠 Usage Examples

1. Auto VoiceOver Labels

import SwiftUI
import AIAccessibilityKit

Image(systemName: "leaf.fill")
 .autoVoiceOverLabel() // AI-generated label via Core ML + Vision

2. Dynamic Font Scaling

Text("Eco Rewards")
    .dynamicFont(.title3) // Automatically scales for accessibility sizes

3. Color Contrast Audit

VStack {
    Text("Accessible Text")
}
.auditContrast(foreground: .primary, background: .white)

4. Touch Target Audit

Button("Continue") { }
    .auditTouchTarget() // Highlights if < 44pt

5. Real-Time Audit Overlay

#Preview {
    AccessibleCard()
        .modifier(AccessibilityAuditOverlay(showGrid: true))
}

Example Component

struct AccessibleCard: View {
    var body: some View {
        VStack(spacing: 12) {
            Image(systemName: "leaf.fill")
                .autoVoiceOverLabel()
                .auditTouchTarget()

            Text("Eco Rewards")
                .dynamicFont(.title3)

            Text("Earn points for sustainable purchases.")
                .dynamicFont(.body)

            Button("Continue") { }
                .padding()
                .background(.blue)
                .foregroundStyle(.white)
                .clipShape(Capsule())
                .autoVoiceOverLabel()
                .auditTouchTarget()
        }
        .padding()
        .background(.white)
        .clipShape(RoundedRectangle(cornerRadius: 16))
        .auditContrast(foreground: .primary, background: .white)
        .modifier(AccessibilityAuditOverlay(showGrid: true))
    }
}

✅ Requirements

• iOS 18+
• Xcode 16+
• Core ML model (e.g., ImageClassifier.mlmodel) for image labeling

📖 How It Works

• Core ML + Vision: Snapshots SwiftUI views, runs OCR and image classification to suggest labels.
• SwiftUI Modifiers: Apply accessibility audits without changing your existing layout.
• Previews Integration: Visualize audits in real-time during development.

🔍 Roadmap

• ✅ Auto VoiceOver labels
• ✅ WCAG contrast audits
• ✅ Touch target checks
• 🔜 Localization-aware AI labels
• 🔜 Snapshot-based accessibility reports for CI/CD

🤝 Contributing Contributions are welcome!

• Fork the repo
• Create a feature branch
• Submit a pull request

📄 License MIT License. See LICENSE for details.