I’m building a small showcase app for a school contest and need a polished Android solution that makes printed text friendlier for students with dyslexia. The flow is straightforward: the user opens the camera, points it at a page, and the app instantly recognises the words (OCR). As soon as the text is captured, each word is spoken aloud in sequence while that same word is highlighted on-screen in real time. The result should feel like a karaoke-style guide that lets the reader follow along word by word without losing their place. Must-have fundamentals • Android only — Kotlin or Java is fine, as long as it runs smoothly on common mid-range phones. • Real-time OCR (Google ML Kit, Tesseract, or any on-device library you prefer) with no dependency on an active internet connection once installed. • Android TextToSpeech API (or equivalent) for clear, natural narration. • Perfect synchronisation between speech and a moving word-level highlight; no line-level or sentence-level fallback. • Simple, clutter-free interface with large, adjustable font and high-contrast highlight colours. Nice touches that will help us in judging – Adjustable speech rate and pitch. – Play / pause / restart controls. – A quick demo mode that ships with a sample paragraph in case no printed material is handy. Deliverables 1. APK ready for installation on Android 8.0+. 2. Complete, well-commented source code. 3. A short README covering build steps, libraries used, and how the word-sync logic works. Acceptance criteria ✔ Captured text is spoken aloud with word-by-word highlighting that tracks the audio in real time. ✔ No internet required during normal use. ✔ App stays responsive on devices with only 3 GB RAM. Once tests confirm these points, the project is done and I can submit it for the contest.