README.md

Gen Slide

Generating PowerPoint slides extracted from a text.

How to run

Set up

Clone the repo:

git clone http://github.com/mehdimo/GenSlide

cd GenSlide

then, create a virtual environment and install packages.

python -m venv ./venv
. ./venv/bin/activate

pip install -r requirements.txt

• Note: Use a Python != 3.9.7 for virtualenv. streamlit won't work in that version. The version here is 3.12.
• We used gpt4ll package to run a LLM server on a local machine. To learn more about gpt4all, see here.

RUN LLM Service

1. Go to llm-service folder and run the gpt.py file.

cd llm-service
python gpt.py

• Running for the first time, the LLM model will be downloaded which may take several minutes.

RUN UI

Navigate to fronend folder and run ui.py using streamlit command:

cd .. 
cd frontend
streamlit run ui.py

It will open the UI in the browser.

Generate PowerPoint slides

• In the UI's textbox, write the text that you want to create presentation based on it. You may use the text in file content.txt in data folder.
• The generated .pptx file will be stored in the frontend/generated folder.

User Interface:

Generated slides in PowerPoint: