You can try using OCR if I understand your task correctly. You have to select the parts of the presentation from which you want to extract the text and use an application such as textsniper.app. You can also use the PDF to Word converter to quickly extract the headings from the original document. But it only converts the whole file and often with errors. So I don't use converters usually. They work well when you need to convert a file from a text format to another text format. Otherwise, there is a lot of inaccuracy.