[mellutussa]

Try https://github.com/ocrmypdf/OCRmyPDF - it uses Tesseract behind the scenes and it absolutely brilliant.