Memory Book

Transform life stories into beautiful illustrated books using AI.

Memory Book is a full-stack web application that turns family memories and reference photos into personalized, AI-generated illustrated memory books. Powered by Google Gemini's multi-agent pipeline, it creates professional-quality books in minutes — not weeks.

Live Demo: memory-book-app-1bfd7.web.app

See how it works

Click the image above to watch the full demo on YouTube

Screenshots

Landing Page

Hero section — "Keep their story close, always" with multi-language support

Impact statistics — Why memory preservation matters

How it Works — 3-step process overview

Sample book showcase with multiple art styles (Watercolor, Cartoon, Anime, Coloring)

Call-to-action section and footer

Authentication

Sign In modal — Google OAuth and email/password authentication

Dashboard

Dashboard — Personal library with example books in different art styles

Settings

Profile settings panel and language selection (PT, ES, EN, DE, FR)

Book Creation Wizard

Choose how to create your book — Detailed story, brief version, or audio (coming soon)

Step 1: Book Setup — Page count, visual reference upload

Step 1: Character description — Gender, skin color, hair color, hair style, accessories

Step 1: Art style selection — Coloring Page, Cartoon, Anime, Watercolor

Step 3: Review & Generate — Final review with reading level and tone settings

Generation Progress

Real-time generation progress — Step-by-step status updates

The Problem

Over 55 million people worldwide live with Alzheimer's or dementia. Every 3 seconds, someone loses a memory. As the disease progresses, the stories that define a person's life — their childhood adventures, first love, proudest moments — begin to fade irreversibly.

Families want to preserve these memories, but the options available today fall short:

Traditional photo albums are static and don't capture the narrative behind each moment.
Professional memory books cost $500–$2,000+ and take weeks to produce.
DIY approaches require design skills, time, and effort most families don't have.

There is no accessible, affordable way to turn a lifetime of stories into a beautiful, illustrated keepsake — until now.

The Objective

Memory Book aims to democratize memory preservation by making it possible for anyone to create a personalized, illustrated memory book in minutes for a fraction of the cost of traditional services.

Core goals:

Accessibility — A simple guided wizard that anyone can use, regardless of technical skill.
Affordability — Reduce the cost from $500+ to under $1 per book using AI generation.
Quality — Produce visually consistent, professionally styled illustrations across every page.
Personalization — Maintain character likeness throughout the book using reference photo analysis.
Global reach — Support multiple languages (English, Portuguese, Spanish, French, German, Italian).

Features

Book Creation Wizard

Step-by-step guided flow to input memories across life phases (childhood, teenage years, adult life, later years).
Upload 1–5 reference photos for character consistency across all illustrations.
Choose from 4 art styles: Watercolor, Cartoon, Anime, and Coloring Book.
Select page count (up to 10 content pages + cover + back cover).
Real-time progress tracking with per-page status updates during generation.

AI-Powered Multi-Agent Pipeline

11 specialized Gemini-powered agents working in coordinated sequence:
- Normalizer Agent — Structures raw user input.
- Narrative Planner Agent — Organizes memories into a coherent story arc.
- Visual Analyzer Agent — Extracts facial and physical features from reference photos (visual fingerprint).
- Prompt Writer Agent — Crafts detailed image generation prompts.
- Prompt Reviewer Agent — Reviews and refines prompts for quality.
- Image Generator Agent — Generates illustrations using Gemini 2.5 Flash Image.
- Quality Validator Agent — Validates output quality with retry logic.
- Additional agents for design review, fixing, and final packaging.
Parallel execution where possible using asyncio.gather().
Quality control loop with up to 3 retry attempts per image.

Interactive Book Viewer

Realistic page-flip book viewer in the browser with spread layout.
Navigate through cover, content pages, and back cover.
Full-screen viewing mode.

PDF Export

Download books as landscape A4 PDFs with professional book-style layout.
Custom fonts embedded (Playfair Display for titles, Lora for body text).
Project logo watermark on each page.
Side-by-side image and text layout with decorative elements.
Cover page, content pages with alternating layouts, and styled back cover.

Dashboard

View all created books in a personal library.
Create new books, delete existing ones.
Track generation progress for in-flight jobs.

Authentication & Storage

Firebase Authentication with Google OAuth and email/password sign-in.
Firebase Firestore for book metadata and user data.
Firebase Storage for reference images and generated illustrations.

Internationalization

Full multi-language support: EN, PT-BR, ES, FR, DE, IT.
Language selector in the navbar.
All UI text, labels, and narratives adapt to the selected language.

Landing Page

Modern, responsive design with animated hero section.
Sample book showcase with interactive page viewer.
Statistics section about memory preservation.
How-it-works guide and privacy/security information.

Tech Stack

Frontend

Technology	Purpose
React 19	UI framework
TypeScript	Type safety
Vite 7	Build tool & dev server
Tailwind CSS 4	Utility-first styling
Framer Motion	Animations & page transitions
React Router 7	Client-side routing
jsPDF	Client-side PDF generation
Firebase SDK	Auth, Firestore, Storage
Lucide React	Icon library

Backend

Technology	Purpose
Python 3.11+	Runtime
FastAPI	REST API framework
Pydantic	Data validation & schemas
Google GenAI SDK	Gemini API integration
Firebase Admin SDK	Server-side Firebase access
Pillow	Image processing & optimization
aiohttp / aiofiles	Async I/O operations

Infrastructure

Service	Purpose
Firebase Hosting	Frontend deployment
Firebase Authentication	User management
Firebase Firestore	NoSQL database
Firebase Storage	File storage (images)
Google Gemini API	AI engine (text, vision, image generation)

Gemini Models Used

Model	Task
`gemini-2.0-flash`	Fast analysis, planning, structured JSON output
`gemini-2.0-pro-exp`	Creative narrative generation
`gemini-2.5-flash-image`	Native image generation (illustrations)

Project Structure

Memorybook/
├── src/                          # Frontend source
│   ├── assets/                   # Images, fonts, sample book
│   ├── components/
│   │   ├── book/                 # Book viewer components (BookViewer, BookEditor, SpreadView)
│   │   ├── layout/               # Navbar, Footer, Dashboard layout
│   │   ├── ui/                   # Reusable UI components (Modals, Cards, Buttons)
│   │   └── wizard/               # Book creation wizard (steps, style selector, upload)
│   ├── constants/translations.tsx # Multi-language translations
│   ├── contexts/                 # React contexts (Auth, Language, Theme)
│   ├── data/                     # Sample book data
│   ├── lib/
│   │   ├── api/                  # Backend API client & hooks
│   │   ├── firebase/             # Firebase config, auth, firestore, storage
│   │   └── generatePdf.ts        # PDF generation logic
│   ├── pages/                    # Route pages (Home, Dashboard, BookPreview)
│   └── index.css                 # Global styles & Tailwind config
│
├── backend/                      # Python backend
│   ├── agents/                   # 11 specialized AI agents
│   ├── clients/                  # Gemini & external API clients
│   ├── models/                   # Pydantic data models
│   ├── pipeline/                 # Pipeline orchestration & validation
│   ├── prompts/                  # Master prompts & language utilities
│   ├── store/                    # Job store & file storage
│   ├── utils/                    # Config, logging, Firebase admin
│   ├── tests/                    # Unit & integration tests
│   ├── app.py                    # FastAPI application entry point
│   └── requirements.txt          # Python dependencies
│
├── public/                       # Static assets (favicons, manifest)
├── firebase.json                 # Firebase hosting & rules config
├── firestore.rules               # Firestore security rules
├── storage.rules                 # Storage security rules
└── package.json                  # Node.js dependencies & scripts

Requirements

Prerequisites

Node.js 18+ and npm 9+
Python 3.11+
Google Gemini API Key (Get one here)
Firebase Project with Authentication, Firestore, and Storage enabled

Environment Variables

Frontend (.env in project root):

VITE_FIREBASE_API_KEY=your_firebase_api_key
VITE_FIREBASE_AUTH_DOMAIN=your_project.firebaseapp.com
VITE_FIREBASE_PROJECT_ID=your_project_id
VITE_FIREBASE_STORAGE_BUCKET=your_project.appspot.com
VITE_FIREBASE_MESSAGING_SENDER_ID=your_sender_id
VITE_FIREBASE_APP_ID=your_app_id
VITE_API_BASE_URL=http://localhost:8000

Backend (backend/.env):

GOOGLE_API_KEY=your_gemini_api_key
HOST=0.0.0.0
PORT=8000
DEBUG=false

Getting Started

1. Clone the repository

git clone https://github.com/juniortorres20/Memorybook.git
cd Memorybook

2. Set up the frontend

npm install
cp .env.example .env
# Edit .env with your Firebase credentials
npm run dev

The frontend will be available at http://localhost:5173.

3. Set up the backend

cd backend
python -m venv venv

# Windows
venv\Scripts\activate

# macOS / Linux
source venv/bin/activate

pip install -r requirements.txt
cp .env.example .env
# Edit .env with your Gemini API key
python run_local.py

The API will be available at http://localhost:8000 with docs at http://localhost:8000/docs.

4. Set up Firebase

Follow the instructions in FIREBASE_SETUP.md to configure authentication, Firestore, and Storage rules.

5. Deploy

npm run build
npx firebase deploy --only hosting

How It Works

User Input → Normalization → Narrative Planning ─┐
                                                  ├→ Prompt Writing → Prompt Review
Reference Photos → Visual Analysis ──────────────┘
                                                          │
                                                          ▼
                                                  Image Generation
                                                          │
                                                          ▼
                                                  Quality Validation
                                                     │         │
                                                  ✓ Pass    ✗ Fail → Fix & Retry (up to 3x)
                                                     │
                                                     ▼
                                              Design Review → Final Book Package

Input — User fills in memories across life phases and uploads reference photos.
Normalization — Raw input is structured and validated.
Planning — A narrative arc is created spanning childhood to later life.
Visual Analysis — Reference photos are analyzed to extract a visual fingerprint (facial features, body type, distinctive traits).
Prompt Creation — Detailed image generation prompts are crafted, embedding the visual fingerprint for character consistency.
Review — Prompts are reviewed and refined for quality.
Generation — Gemini 2.5 Flash Image generates each illustration.
Validation — Each image is validated; failed images are regenerated up to 3 times.
Delivery — The complete book is packaged and made available for viewing and PDF download.

API Documentation

The backend provides a RESTful API. Full endpoint documentation is available at /docs when running the server.

Method	Endpoint	Description
`POST`	`/jobs`	Create a new book generation job
`GET`	`/jobs/{job_id}`	Get job status and progress
`GET`	`/jobs/{job_id}/result`	Get completed book data
`GET`	`/jobs/{job_id}/assets`	List all job assets
`GET`	`/assets/{job_id}/{folder}/{filename}`	Serve an asset file
`GET`	`/jobs`	List recent jobs
`DELETE`	`/jobs/{job_id}`	Delete a job and its assets
`GET`	`/languages`	Get supported languages
`GET`	`/health`	Health check

See backend/README.md for detailed request/response examples.

Scripts

# Frontend
npm run dev          # Start development server
npm run build        # Production build
npm run preview      # Preview production build
npm run lint         # Run ESLint

# Backend
python run_local.py  # Start backend server
pytest               # Run tests
pytest -v --cov=.    # Run tests with coverage

# Firebase
npx firebase deploy --only hosting              # Deploy frontend
npm run firebase:deploy:rules                    # Deploy Firestore & Storage rules
npm run firebase:emulators                       # Start Firebase emulators

License

This project is licensed under the MIT License. See LICENSE for details.

Name		Name	Last commit message	Last commit date
Latest commit History 11 Commits
.agent/rules		.agent/rules
backend		backend
public		public
src		src
.env.example		.env.example
.firebaserc		.firebaserc
.gitignore		.gitignore
LICENSE		LICENSE
README.md		README.md
cors.json		cors.json
eslint.config.js		eslint.config.js
firebase.json		firebase.json
firestore.indexes.json		firestore.indexes.json
firestore.rules		firestore.rules
index.html		index.html
package-lock.json		package-lock.json
package.json		package.json
postcss.config.js		postcss.config.js
storage.rules		storage.rules
tailwind.config.js		tailwind.config.js
tsconfig.app.json		tsconfig.app.json
tsconfig.json		tsconfig.json
tsconfig.node.json		tsconfig.node.json
vite.config.ts		vite.config.ts

Folders and files

Latest commit

History

Repository files navigation

Memory Book

See how it works

Screenshots

Landing Page

Authentication

Dashboard

Settings

Book Creation Wizard

Generation Progress

The Problem

The Objective

Features

Book Creation Wizard

AI-Powered Multi-Agent Pipeline

Interactive Book Viewer

PDF Export

Dashboard

Authentication & Storage

Internationalization

Landing Page

Tech Stack

Frontend

Backend

Infrastructure

Gemini Models Used

Project Structure

Requirements

Prerequisites

Environment Variables

Getting Started

1. Clone the repository

2. Set up the frontend

3. Set up the backend

4. Set up Firebase

5. Deploy

How It Works

API Documentation

Scripts

License

About

Topics

Resources

License

Uh oh!

Stars

Watchers

Forks

Contributors

Uh oh!

Languages