Ah nice, one thing I would have done though is use phonemes to more precisely control the pronunciation. There are a lot of places where it could have been useful there to have that done. Though when you do that, the problem then becomes using the app, if you write the files out with the app, they're sampled at 11K and it just doesn't sound as good. SO you have to record them externally and use the app to play it.
The timing has to be the most challenging part I'd imagine, I've never attempted this, or anything like it before, but I have a way I might approach it. I would probably import all the files into reaper on the same track, so I would have multiple items. I would then start aligning them, and splitting off syllables where needed, and moving them. I would probably enter the correct time signature and tempo into the project before importing the music track so that the reaper metronome would run correctly, and use that to help me place on beat. If I It would be very time consuming, and I generally don't do projects like that, not because I wouldn't enjoy it, but because I get way too obsessive over it. Even just editing normal audio, I might spend 5-10 minutes on a 100ms segment trying to get it to sound right, like preserving the cadence of the speech after I have made a change to the surrounding speech. So editing and stuff like that takes me far far longer than it should because I can't ever let anything go.
I'm a cat! What's mine is mine, and what's yours is mine to :P XD