Skip to content
All projects
Utilities2024AI-powered

AI Photo Translator & Scanner

Point the camera at any text and read it in your language.

  • iOS
Vision
Recognition
Realm
Store
Camera + PDF
Input

About

An iOS app for real-time photo translation and text extraction. The Vision framework finds and reads text in the camera feed, the Google Translation API renders it in the target language, and an AI tutor layer lets people ask follow-up questions about what they have just translated. It also handles document and PDF text extraction.

What shipped

  • Real-time camera translation using the Vision framework
  • Google Translation API integration across many language pairs
  • PDF and image text extraction pipelines
  • A new Realm data model for saved translations
  • AI tutor prompts and chat for an interactive learning loop

How it was built

  1. Recognition

    Vision framework text detection over a live camera feed.

  2. Translation

    Google Translation API across a wide set of language pairs.

  3. Extraction

    Document and PDF text extraction pipelines feeding the same translator.

  4. Tutor layer

    AI chat over the translated text, turning a lookup into a learning loop.

Challenges & solutions

  • ProblemA translation you cannot ask questions about teaches you nothing.

    SolutionAn AI tutor layer sits on the translated text, so a phrase can be unpacked rather than just converted.

Built with

  • Swift
  • UIKit
  • Vision
  • Google Translation API
  • Realm