I've seen some Vocaloid style stuff with Aquestone and UTAU. (The following apps being more inexpensive or free Vocaloid clones.) I'm guessing that's what you're looking for.
Unfortunately I find the English documentation for these apps very lacking, thus for me they're somewhat useless at this point. I managed to get Aquestone to work with LMMS. Currently it only supports a Japanese phenome set - I couldn't get much more than "banana" out of it as far as words that I actually understand in English. (Although if I want my computer to sing an ode to a banana... ) I figure this is pretty much the case for other languages as well.
For UTAU, I can't find much more than some occasional talk about it and vids on YouTube.
Of course in time we may see translations and phenome sets for non-Japanese languages regarding these apps. But at this moment? Not that I'm aware of. Still, it doesn't hurt to keep an eye on them for new developments that may make them more useful. The concept and execution is there, but there's not much addressing the greater world audience wanting these apps.
The best I can do is take the "wait and see approach".
-----Edit-----
You can scratch Utau off the list as being a plug-in. I found more on it, and it seems to be it's own program which is a midi-tracker of sorts.
As a last resort you can sort of make speech synthesis/dictation like MS-Sam sing, but it's very labor intensive recording and chopping up words and phrases with software like Audacity. And each non-repeated word/syllable would be it's own sample track. Such a process is crazy, but not impossible. (AFAIK)