An mp3 file will be returned... you should have a player (such as WinAMP) which understands .m3u -- streaming mp3s.
What is going on here?
The text you enter is split into words. The words are then broken down into phonemes. There are about 44 phonemes in the English language. A phoneme is a sound smaller than a syllable, for instance, here is a translation from word-->phonemes:
MONKEY = M AH NG K IY
The word monkey has 5 phonemes.
I found a phoneme dictionary on the Carnegie Mellon site, recorded my voice saying all the phonemes, and wrote a script to turn mush all the .wav files into an mp3. I used bladeenc to encode the .wav files into the mp3 format.