AudioLens 🎧📖

A dyslexia-friendly web app that uses local AI to perfectly sync audiobooks with PDFs.

---

🌟 Overview

AudioLens is an offline-first, privacy-focused reading companion designed to make reading more accessible. It seamlessly pairs your PDFs with audiobooks, utilizing OpenAI's Whisper model (running entirely on your machine) to generate word-level timestamps. The result is a synchronized, highlighted reading experience where the text glows block-by-block exactly as it is spoken.

✨ Features

• 🧠 Local AI Transcription: Uses the Whisper tiny model to transcribe and accurately timestamp audiobooks completely offline. No API keys required, no data leaves your machine.
• 🎯 Word-for-Word Sync: Enjoy precise, 60fps caption-style highlighting synced directly to the audio track.
• 📚 Smart Dashboard & Resume: Automatically saves your uploaded books and playback progress to your browser (localStorage), resuming exactly where you left off.
• 🛠️ Adjustable Split-View: A premium, glassmorphism dark-mode UI with a draggable divider lets you resize the PDF and transcript panels exactly to your liking.
• 🛡️ Auto-Repairing PDFs: Built-in PDF structural repair (pikepdf) ensures that damaged or malformed PDFs load perfectly in the browser.
• ⚡ Content-Addressable Storage: Smart file deduplication prevents wasting disk space if you upload the large audiobooks multiple times.

🚀 Installation

Ensure you have Python 3.8+ and ffmpeg installed on your system.

1. Clone the repository:

   git clone https://github.com/yourusername/audiolens.git
   cd audiolens

2. Run the setup script: The included run.sh script will automatically create a virtual environment, install the necessary dependencies, and start the local server.

   ./run.sh

3. Open the App: Navigate to http://127.0.0.1:8000 in your web browser.

📁 Project Structure

• backend/main.py: The FastAPI server handling uploads, Whisper transcription, background caching, and PDF repairing.
• frontend/: Contains the vanilla HTML, CSS, and JS powering the premium split-view interface.
• uploads/: Where your files, transcript caches (.json), and database (jobs.json) are stored locally.

🤝 Contributing

Contributions, issues, and feature requests are welcome!

📝 License

This project is open-source and available under the MIT License.