wolketich/mcq-ai

Answer multiple choice question through ChatGPT-3.5

PythonMIT

mcq-ai

Answer multiple choice question through ChatGPT-3.5

OCR Multioption with ChatGPT

Convert images to text using OCR and process the extracted text with ChatGPT.

Overview

This application allows users to:

Convert images containing text into plain text using Optical Character Recognition (OCR).
Beautify the extracted text using OpenAI's GPT-3.5 Turbo Instruct model.
Process the text with ChatGPT to get answers or further insights.

Features

Image Input: Accepts images from both disk and clipboard.
Text Beautification: Enhances the readability of the extracted text.
Interactive Mode: In debug mode, users can choose various options like printing the text, saving it to a file, or processing it with ChatGPT.

Requirements

Python 3
PIL (Python Imaging Library)
pytesseract
openai

Setup

Clone the repository:

git clone https://github.com/wolketich/mcq-ai/

Navigate to the project directory:
```
cd mcq-ai
```
Install the dependencies:
```
pip install -r requirements.txt
```

Set up the config.json file with your OpenAI API key:

 {
     "openai": {
         "api_key": "YOUR_API_KEY"
     }
 }

Usage

Run the application:
```
python3 app.py
```
or run the application in debug-mode that allows interactive options:
```
python3 app.py --debug
```