I'm actually in the finishing stages of making a PHP Hyphenator class based upon Frank Liang's algorithm and the TeX dictionaries, which pretty much seems to be the appoach taken by all office suites. (Actually I found this topic while looking for a good name for it that wasn't already taken). With slowly improving support from browsers for the entity, it's becoming a realistic option to hyphenate content in websites.
Core functionality is working; splitting (and thus counting) and/or hyphenating text and/or HTML, parsing TeX hyphen dictionaries, caching those parsed dictionaries. Some planned features are still missing but nothing that stops you from using it. Also there's no good documentation, samples, formal unittest or vanity site yet.
I've created a github site for it here and will post the current version on it ASAP, so check back in a few days.
I've only tested it with Dutch (my native language) and US English, so it may still have some issues with languages using different character sets.
