PyDocX is a parser that breaks down the elements of a docxfile and converts them into different markup languages. Currently, only HTML is supported. Markdown and LaTex are planned for the future. You can extend any of the available parsers to customize it to your needs. You can also create your own class that inherits DocxParser to create your own methods for a markup language not yet supported.
To get started using PyDocX, see the Usage guide and also Extending PyDocX.