This is the Windows app named Savvy DOCX Recovery whose latest release can be downloaded as savvy_corrupt_DOCX_setup_3.0.0_without_adware.exe. It can be run online in the free hosting provider OnWorks for workstations.
XML was designed from the beginning to be intolerant of errors. This decision adversely affects MS Word's corruption recovery. With one error in the document.xml subfile where all the DOCX file's text is stored, instead of a partial recovery, Word will stop and throw an error.
Savvy DOCX Recovery attempts to do precise surgery on corrupt Word documents to reorder or excise bad XML tags.
If this doesn't work, it uses the command line app xmllint first to attempt to repair corrupt XML subfiles of the DOCX. If this doesn't work a second attempt is made where the corrupt XML subfiles are truncated at the first error, and the correct ending tags are again added with xmllint.
If all else fails, SilverCoder's DocToText is used to extract text.
Try also and -us/kb/2528942 and my other SF projects: Corrupt Extractor for Microsoft Office, Corrupt DOCX Salvager, S2 Recovery Tools for Microsoft Word.
This program is called "Savvy DOCX Recovery". It is automatically recovers corrupt docx files. The program automatically uses four different progressively more basic algorithms until the final method hopes only to recover text, no formatting.
First thing to remember is that Office uses the Office XML format. Every .docx document is basically a zip file. The content is stored within, including the document metadata, structure and even some SharePoint metadata.
Now save the file again, place it back into the .zip file. Rename the .zip file to .docx and try to open it. In my case, I got another error message. This one also to do with . I removed that section as well.