The Intention with this application is to act as a proof of concept that we can randomly generate unique stories based on an uploaded .jpeg file.
The idea came about after family members remarked how hard it was to get children's books at the library for my nephew.
Ideally, after future iterations it could seamlessly generate short stories that are appropriate for someone his age.
Also, I wanted to try out web dev, thus the choice of using Flask and boilerplate HTML with some JS for async and backend communication
As of writing, it is not hosted publically, so to run the app you would need to clone the repo locally. Also, be sure to have the necessary libraries installed:
pip install Flask
pip install openai
To protect your OpenAI API key, set up the necessary environment variables by creating a .env file in the root directory
i.e OPENAI_API_KEY=your-openai-api-key
Make sure the static/uploads directory exists to store uploaded images. You can manually create it with: mkdir -p static/uploads
Run the Application with flask by inputting flask run at the command line, and the local server will be launched
I want to make this program more feature-complete, this will involve:
ability to fine tune narrative text based on reading level
incorporating AI image generation in combintion with text generation to remove the need for uploading and become one-button operational
A toggle for genres and/or flavour text
Troubleshooting .png image previewing and optimizing file storage