Tourist Assistant

An interactive voice-enabled tourist guide that provides information about cities, landmarks, and destinations worldwide. This application uses OpenAI's GPT models for text generation and speech features for a natural conversation experience, along with RAG capabilities and Google Places API integration for real-time attraction information.

Features

Text-based chat interface for asking questions about tourist destinations
Voice input capability through microphone recording
Audio responses using OpenAI's text-to-speech technology
Clean, responsive user interface with Gradio
RAG (Retrieval-Augmented Generation) system using PDF knowledge base
Google Places API integration for real-time information about attractions
Set current location for contextual queries
Quick access to nearby attractions information

Requirements

Python 3.9+
OpenAI API key
Google Places API key (optional, for location search features)

Installation

Clone this repository
Install the required dependencies:
```
pip install -r requirements.txt
```

Create a .env file in the project directory with your API keys:

OPENAI_API_KEY=your_openai_api_key_here
GOOGLE_PLACES_API_KEY=your_google_places_api_key_here

(Optional) Add PDF files to the knowledge-base/ directory to enhance the assistant's knowledge about specific locations

Running the Application

Start the application by running:

python tourist-assistant.py

The interface will automatically open in your default web browser. If it doesn't, navigate to the URL shown in the terminal (typically http://127.0.0.1:7860/).

Usage

Type your question about any tourist destination in the text box
Or click the microphone button and speak your question
The assistant will respond with text and spoken audio
Set your current location using the "Set Location" feature
Click "Nearby Attractions" to get information about attractions near your current location
Use the "Refresh Knowledge Base" button to reload PDFs in the knowledge-base directory
Use the "Clear" button to start a new conversation

Technologies Used

OpenAI GPT-4o Mini for chat completions
OpenAI Whisper for speech-to-text
OpenAI TTS for text-to-speech
Langchain for RAG implementation
FAISS for vector storage
Google Places API for location-based attraction information
Gradio for the web interface
pydub for audio processing

2.5 KiB Raw Permalink Blame History