🎧 PDFtoAudio — Accessible AI Learning & Document Platform

Turn any document into an accessible lesson you can listen to, understand, and remember.

PDFtoAudio is an accessible AI learning & document audio studio. Convert PDFs, DOCX, & eBooks into neural audiobooks, 2-host AI podcasts, sentence-synced readers, flashcards, quizzes, and grounded Q&A chat with page citations.

---

🌟 Vision & Positioning

PDFtoAudio transforms PDFs, textbooks, research papers, and documents into interactive audio lessons, grounded Q&A study tools, and customizable revision materials.

---

✨ Core Feature Highlights

📚 1. Document Library & Resume Playback

• Multi-Format Upload: Upload PDFs, DOCX, TXT, Markdown, and EPUB files.
• Metadata Index: Total page count, word count, estimated listening time, and parser quality reports.
• Resume Playback: Automatically saves your exact sentence position and playback timestamp.
• Bookmarks & Notes: Save highlights, notes, and sentence bookmarks.

🧠 2. Educational Study Suite

• 6 Summary & Explanation Modes:
  • 📋 Quick Overview
  • 🔒 Strict Read Mode (Narration without AI paraphrasing)
  • 🔬 Educational Deep Dive
  • 💡 Explain Simply (ELI5)
  • 🎓 Professor Analysis
  • ⚛️ Feynman Learning Technique
• Interactive Revision Flashcards: Click-to-flip flashcards with difficulty levels and page-number citations.
• Anki CSV Export: Export flashcards directly to .csv format for Anki spaced-repetition software.
• Multiple-Choice Quizzes: Interactive assessments with instant score feedback and explanatory citations.

💬 3. Trust Mode & Document Chat

• Grounded Document Q&A: Ask any question about your document.
• Page-Level Citations: Answers are strictly grounded in document text with clickable page-number references (Page 2, Page 5).
• Unverified Fallback: Clearly displays "Not found in document" when evidence is missing.
• Text-to-Speech Answer Conversion: Listen to any chat response instantly.

♿ 4. Accessibility Mode (WCAG 2.1 AA)

• Dyslexia-Friendly Font: Full built-in support for OpenDyslexic.
• High Contrast Theme: High-visibility dark theme for visually impaired users.
• Text Scaling: Normal, Large (120%), and Extra Large (150%) display scaling.
• Full Keyboard Navigation: Visible focus outlines (:focus-visible) and complete keyboard controls:
  • Space / K: Play / Pause
  • M: Mute / Unmute
  • ?: Accessibility Drawer
• Sentence & Word Karaoke Highlighting: Visual sentence highlighting synced with audio playback.

🎙️ 5. Neural Audio & Podcast Engine

• 50+ Neural AI Voices: High-definition voices across English, Hindi, Bengali, Spanish, French, German, Japanese, etc.
• 2-Host AI Podcast Mode: Synthesizes an episode featuring Alex (Male Storyteller) and Sam (Female Studio Host) with page citations.
• Pronunciation Dictionary: Custom replacements for technical terms, acronyms, and formulas.
• RSS Podcast Feed: Generate valid RSS 2.0 XML feeds (/api/documents/{id}/rss) for Apple Podcasts, Spotify, and Pocket Casts.

---

🛠️ Architecture & Backend Services

backend/
├── app.py              # FastAPI Web Application & REST API routes
├── parser_service.py   # Multi-format document parser & Quality Reports
├── ocr_service.py      # Scanned PDF detection & Tesseract OCR pipeline
├── storage_service.py  # Local persistent Library CRUD, Metadata & Bookmarks
├── tts_engine.py       # Neural TTS Engine (edge-tts + gTTS + Pronunciation Dictionary)
├── podcast_engine.py   # 2-Host AI Podcast Generator with Citations
├── summary_service.py  # Educational Summary Engine (6 Modes)
├── quiz_service.py     # Flashcards, Quizzes & Anki CSV Exporter
├── chat_service.py     # Grounded Q&A Chat Engine with Page Citations
├── export_service.py   # Subtitles (.VTT/.SRT) & RSS Podcast Feed Generator
├── task_service.py     # Non-blocking async background job queue
└── schemas.py          # Pydantic data schemas

---

🚀 Quick Start

1. Install Python dependencies:

   pip install -r requirements.txt

2. Launch the Application:

   python main.py

   This boots the server at http://127.0.0.1:8000 and opens your browser.

3. Run Automated Test Suite:

   python3 -m unittest discover tests

---

🔌 API Reference

Method	Endpoint	Description
GET	/	Web Studio Application
GET	/api/voices	Returns list of available AI voices
POST	/api/upload	Upload document to Library & extract metadata/chapters
GET	/api/documents	List all documents in Library
GET	/api/documents/{id}	Get metadata and content for a document
POST	/api/documents/{id}/playback	Save last played sentence index & timestamp
DELETE	/api/documents/{id}	Delete document from Library
POST	/api/documents/{id}/summary	Generate educational summary across 6 modes
POST	/api/documents/{id}/study	Generate flashcards and quiz questions
GET	/api/documents/{id}/anki-csv	Download flashcards in Anki CSV format
POST	/api/documents/{id}/chat	Grounded Q&A chat with page citations
GET	/api/documents/{id}/rss	Download RSS Podcast Feed XML

---

📄 License & Security

Distributed under the MIT License. Developed by Codexia_afk. See SECURITY.md and ACCESSIBILITY.md for data protection and conformance statements.