Yah, I have actually been in conversation with the Flite author recently. I compiled Flite last night and it does a good job of creating WAV files of the spoken text, but I can't get it to output sound because of the 4608 byte restriction on /dev/audio. Flite writes 256 bytes at a time. I'm going to try to hack around it, the author wasn't able to give me any help on that front, unfortunately.

I was just asking if Kim was using Flite or something else. If he had gotten Flite to output to /dev/audio, he would have solved one of my current problems and I would have begged to see what he did. :)

Interestingly enough the Flite package is being rolled into a commercial product. The author says that Flite will be back-filled with the innovations from the commercial product, but the voices won't be as good on Flite to encourage people to buy it. Whenever I hear this, I translate that into "the freeware/open-source product is now dead, forget about it." I am sure the Empeg will be an exception to that rule. :) But he says that Flite will get better voices and better part-of-speech recognition soon. Until that happens, I guess PCM is the way to go.