Voyara：基于 Google Gemini 的 AI 旅行规划助手

AI-Powered Travel Planning Assistant

Voyara is an AI-powered travel planning assistant that turns your destination, budget, duration, travel preferences, and starting location into a practical, personalised trip plan.

It combines Google Gemini, a Flask backend, interactive maps, destination discovery, and a responsive web interface into one travel-planning experience.

🌍 Live Demo

🚀 Try Voyara Live

📂 View Source Code

✨ Features

🤖 AI Trip Planning

Generate personalised day-by-day itineraries.
Plan trips using destination, duration, budget, traveller count, starting location, travel month, and preferences.
Create practical time-block plans for shorter trips.
Get concise, travel-focused recommendations.

🔄 Modify & Regenerate

Modify an existing itinerary using natural-language instructions.
Regenerate an alternative itinerary while keeping the original trip requirements.
Refine your trip without starting from scratch.

🗺️ Explore Destinations

Search destinations using OpenStreetMap/Nominatim.
Explore locations on an interactive Leaflet map.
Open destination details with AI-generated travel guidance.
Start planning directly from an explored destination.

📍 For You Recommendations

Discover places based on your interests and selected category:

🍴 Food & Cafés
🏛️ Attractions & Culture
🌿 Nature & Outdoors
🎯 Experiences & Activities
🛍️ Shopping & Local Markets

Voyara supports Google Places when configured and uses an OpenStreetMap-based fallback for place discovery.

💬 Voyara Assistant

An integrated AI travel assistant that can help with:

Trip planning
Budgeting
Packing
Destination questions
Travel guidance
App guidance
Travel safety

The assistant also keeps recent conversation context during a session.

💾 Save & Manage Trips

Save selected travel data locally in the browser.
Keep generated trip information available for later use.
Print itineraries in a PDF-friendly format.

📱 Progressive Web App

PWA support with a service worker.
Responsive interface for different screen sizes.
Browser Local Storage for selected travel data.

🧠 AI & Backend

Voyara uses the Google Gemini API through the `google-genai` Python SDK.

The Flask backend provides APIs for:

✈️ Trip planning
🔄 Itinerary modification
♻️ Itinerary regeneration
📍 Destination guides
💬 AI assistant chat
🗺️ Place discovery
❤️ Health/status checks

The backend also includes retry and fallback handling for temporary Gemini API failures.

🛠️ Tech Stack

Frontend

HTML5
CSS3
JavaScript
Leaflet.js
Progressive Web App APIs
Browser Local Storage

Backend

Python
Flask
Flask-CORS
Google GenAI SDK
python-dotenv

APIs & Services

Google Gemini
Google Places API (when configured)
OpenStreetMap / Nominatim
Leaflet

Deployment

Render — frontend & backend hosting

🏗️ How It Works

User Preferences
      ↓
Voyara Frontend
      ↓
Flask Backend
      ↓
Google Gemini + Location Services
      ↓
Personalised Travel Results
      ↓
Interactive Itinerary & Destination Experience

📁 Project Structure

Voyara/
├── frontend/
│   ├── index.html
│   ├── style.css
│   ├── app.js
│   └── service-worker.js
├── main.py
├── requirements.txt
├── .gitignore
└── README.md

⚙️ Run Locally

1. Clone the repository

git clone https://github.com/upasana-s7/Voyara.git
cd Voyara

2. Create a virtual environment

Windows PowerShell:

python -m venv .venv
.\\.venv\\Scripts\\Activate.ps1

3. Install dependencies

pip install -r requirements.txt

4. Configure environment variables

Create a `.env` file in the project root:

GEMINI_API_KEY=your_gemini_api_key
GOOGLE_MAPS_API_KEY=your_google_maps_api_key

Never commit real API keys to GitHub.

5. Start the backend

python main.py

The Flask backend runs locally on:

http://127.0.0.1:5000

6. Open the frontend

Open:

frontend/index.html

For complete AI functionality during local development, keep the Flask backend running.

🔐 Environment Variables

Variable	Purpose
`GEMINI_API_KEY`	Required for Gemini-powered itinerary generation, modification, regeneration, destination guidance, and assistant chat
`GOOGLE_MAPS_API_KEY`	Optional; enables Google Places-based recommendations when configured

Voyara uses an OpenStreetMap-based fallback for destination and place discovery where applicable.

🚀 Deployment

Voyara is deployed as two connected services on Render:

Frontend

`https://voyara-frontend.onrender.com\`

Backend

`https://voyara-backend-cwik.onrender.com\`

The frontend communicates with the deployed Flask backend through the configured API base URL.

API keys remain server-side and are not exposed in the frontend.

🎯 Project Highlights

Voyara brings multiple travel-planning tasks together in one application:

Personalised Planning
Generate itineraries around real trip constraints and preferences.

AI-Powered Refinement
Modify and regenerate plans using natural language.

Destination Discovery
Explore destinations, places, and attractions through an interactive map.

Smart Recommendations
Discover food, attractions, nature, experiences, and shopping based on category.

AI Travel Assistance
Ask questions and get travel-focused guidance inside the application.

Offline-Friendly Data
Use browser storage to retain selected travel information.

💡 Why Voyara?

Planning a trip often means switching between multiple tools for:

Finding destinations
Researching places
Building an itinerary
Checking activities
Adjusting plans
Getting travel advice

Voyara brings these steps together into a single AI-powered travel experience.

👩‍💻 Author

Upasana S

🔗 GitHub: https://github.com/upasana-s7

📌 Project Status

Completed personal AI travel-planning project.

Voyara is available as an open-source project on GitHub with a live deployed demo.