Extracting embedded word pad from Microsoft Word document (docx) using java

Question

I have a word document which has multiple OLE objects embedded into it(MS word,Excel,PowerPoint and Word Pad) MS word exposes all other files and images as they are but it exposes the word pad and charts as binary files. I want to know if there is any tool available with Java APIs to extract or if it is possible to write some custom code by myself to convert these files back to RTF.