Fixed maximum character allowance and completely refactored speech generation to accomplish this

Added option to start playback from any position on a page (right click on a word and select "Start reading aloud from here" from the context menu)

Added optional keyboard shortcuts for reading the page, a subsection, etc.

Improved CSS handling so it doesn't potentially conflict with other websites

Fixed bug relating to the text-to-speech API server having new, ambiguous requirements

Hopefully this version fixes the issue, but I will continue monitoring the situation