In the document.xml component of a Word document, each segment of the text is preceded by <w:t> tag and the text string is terminated by another tag that would normally be </w:t> In the case of the xml code removed from the document.xml component to allow the document to be opened, it was not always </w:t> that followed the text string, rather another tag as all tags begin with the < character, I have run a macro on the code that was removed and pasted into a Word document (of 1536 pages of single spaced code) to extract each of the </w:t> tags and the characters following each one upto the next "<" character and insert it into a new document.
This is the result of running that macro
<w:t>e
<w:t>a.
<w:t>a.
<w:t>iˈxaɀ"Arial"/>
<w:t>.
<w:t>e
<w:t>a.
<w:t>a
<w:t>a
<w:t>n
<w:t>gu
<w:t>a
<w:t>a
<w:t>Á
<w:t>iˈnal
<w:t>iˈnal
<w:t>iˈnal
<w:t>⟨
<w:t>a
<w:t>a
<w:t>a
<w:t>iˈnal
<w:t>a
<w:t>a
<w:t>a
<w:t>a
<w:t>a
<w:t>n
<w:t>'pɾin
<w:t>a.
<w:t>iˈnal
<w:t>gu
<w:t>gu
<w:t>gu
<w:t>gu
<w:t>gu
<w:t>IPA
<w:t>gu="00456CF3">
<w:t>IPA
<w:t>'pɾin
<w:t>a.
<w:t>e
<w:t>iˈxaɀ"Arial"/>
<w:t>.
<w:t>iˈxa>
<w:t>jar
<w:t>e
<w:t>a
<w:t>ˈa.
<w:t>a >
<w:t>jar
<w:t>8 r>
<w:t>8 r>
<w:t>a
<w:t>a
<w:t>a
<w:t>a
<w:t>a
<w:t>jar
<w:t>jar
<w:t>jar
<w:t>jar
<w:t>jar
<w:t>a >
Nothing more can be recovered from the file.