Still a little choppy, but the timing is better than my first attempt. I was hoping that adding fake V C would make it sound smoother, but it's probably better to avoid the extra work.
I don't plan on doing anything more with this voice bank.
I converted an UTAU CV voice bank to DeepVocal just for fun (plus I needed some practice using the toolbox). It sounded better than I thought it would. The timing is slightly off, but I think it's possible to have a decent DV voice bank that only uses CV samples.