Lingua::EN::Syllable::syllable() estimates the number of syllables in the
word passed to it.

Note that it isn't entirely accurate... it fails (by one syllable) for
about 10-15% of my /usr/dict/words. The only way to get a 100% accurate
count is to do a dictionary lookup, so this is a small and fast alternative
where more-or-less accurate results will suffice, such as estimating the
reading level of a document.

I welcome pointers to more accurate algorithms, since this one is pretty
quick-and-dirty. This was designed for English (well, American at least)
words, but sometimes guesses well for other languages.