Hmmm.... Working on a few lip syncing tests. Getting mouth motion to match up to dialog is actually pretty easy to automate, but surprisingly hard to make look natural. I'm basically loading up the audio and using it to drive lip movement, which works but I think the audio needs to be pre-smoothed to get it to look less abrupt.
Once I've got that worked out I'm going to start rendering out shots for the first scene.
It looks like this scene is going to work out to be a little over 3 minutes, so in terms of render time... that's only about 5400 frames. Seems reasonable that I'll be done by the end of August.