Unifies, cleans and converts NotebookLM presentations in a single step.
This script automates the entire NotebookLM presentation cleaning process:
- Merges multiple PPTX files into one
- Converts to lossless PDF
- Removes NotebookLM watermark
- Converts back to clean PPTX
NotebookLM generates presentations with a maximum of 15 slides per file. For longer documents, you end up with multiple PPTXs that need to be:
- Merged into one
- Converted to PDF
- Cleaned from watermark
- Converted back to PPTX
This script does ALL of that with a single command.
# Clone the repository
git clone https://github.com/Milor123/notebooklm-unified.git
cd notebooklm-unified
# Install dependencies (or the script will auto-install them)
pip install -r requirements.txt# Process all PPTXs in current directory
python workflow.py
# Specify input folder
python workflow.py --input "C:\Downloads\presentations"
# Custom output filename
python workflow.py --output "my_clean_presentation.pptx"
# Debug mode (shows watermark coordinates)
python workflow.py -d| Option | Description |
|---|---|
-i, --input |
Input folder or PPTX file |
-o, --output |
Output filename (default: presentacion_limpia.pptx) |
-v, --verbose |
Verbose mode |
-d, --debug |
Debug: shows watermark coordinates |
# Let's say you have:
# part1.pptx (15 slides)
# part2.pptx (15 slides)
# part3.pptx (10 slides)
python workflow.py --output "complete_presentation.pptx"
# Result: complete_presentation.pptx (40 slides, no watermark)- Python 3.8+
- Windows, macOS or Linux
Dependencies are automatically installed the first time you run the script.
PPTXs (multiple)
→ Merge PPTXs (extract images, maintain order)
→ PDF lossless (img2pdf)
→ Remove watermark (column-by-column sampling algorithm)
→ Clean PPTX (PyMuPDF + python-pptx)
- Watermark Remover: Algorithm based on work by neosun100/notebooklm-watermark-remover
- Libraries: python-pptx, img2pdf, PyMuPDF
MIT License - see LICENSE
⭐️ If this script was useful to you, consider giving the project a star!
