I guess you're trying to publish a book as mobi ebook or something like that?

I don't know that there's a good solution to this problem. What you can do to make the HTML code as clean as possible is to use consistent formatting in Word throughout the document.

Meaning you use the Styles on the Home tab for all your formatting, instead of manually changing font-size, font-style, or paragraph attributes like line height. You can add custom ones, and you should delete all you don't need.

You will still have to clean up the generated HTML, but it should be a good deal less messy.

just use wordpress websites