Check out Allen Institute for AI’s online tool Paper to HTML Converter, an experimental prototype that aims to render scientific papers in HTML so they can be more easily read by screen readers or on mobile devices.

Still in beta, the system is currently able to process PDFs, LaTeX source, and PubMed Central XMLs. To use this system, upload a scientific PDF (or LaTeX zip file or JATS XML) on the main page. When you press upload, the system will process the document in the background and return the HTML when it is done. This usually takes around 1-2 minutes per PDF depending on the size of the file, and is much faster for LaTeX or XML.

Note that there are still several known limitations when using the system. Tables are currently extracted from PDFs as images, which are not accessible. Mathematical content is either extracted with low fidelity or not being extracted at all from PDFs. Processing of LaTeX source and PubMed Central XML may lack some of the features implemented for PDF processing. The folks at Allen institute for AI are working to improve these components.

You can also have a look at the Paper Example Gallery. Clicking on one will render a reader-friendly HTML version.