You might
want such thing. But you won't get any.
A PDF is a "document" format, but it isn't really one. It is a special encapsulated PostScript file, which is meant to be a printing standard. Thus a PDF file is built to be rendered on a printing device, not to be edited as text. You might find a word having it's letters in the file as independent objects semantically not related to each-other in any form.
So your only chance is to:
1) render the PDF to an image
2) process the image with an OCR tool
3) convert the result to word.
If you want open source / free solutions you might find some independent tools for these steps.
But if you don't have the time and/or the knowledge to get those components working together, I suggest you look for a commercial solution covering all steps. I suggest you look at this one:
http://www.abbyy.com/ocr_sdk_windows/technical_specifications/[
^]