First try selecting a sentence in the PDF. If selection and copy work, ordinary PDF-to-text extraction may be enough. If each page is just an image, you need optical character recognition, or OCR, rather than a text extractor alone.
Use the clearest scan available: upright pages, adequate resolution, even lighting, and minimal background clutter. After OCR, check headings, tables, names, amounts, and page order against the original. OCR can silently confuse similar characters such as 0 and O, so do not use its output for important figures without review.
Keep the source PDF until the extracted text has been verified.
Akousa