If I'm going to work on the vocals, I'll micro edit everything.

That means going through the entire vocal, slicing the vocal, and adjusting the balance of everything. My DAW (Reaper) creates crossfades at split points, so they splits are (mostly) seamless. As long as the split isn't in the middle of a sustained vowel, it's not going to be heard.

By doing this, I can do several things at the same time:

  • Balance the vocals so everything is at a constant volume. No compressor needed.
  • Adjust consonants. No de-esser needed.
  • Fix timing issues (notes too long/to short/not on beat)


It's a bit tedious, but once I'm in the groove, it goes quickly. Plus, I get to closely review the vocal, which is the most important part of the track.

AI voices require a bit less attention, but I try to make at least one pass through to make sure the consonants are clear. A big part of clear vocals comes from being able to hear the consonants.

Plus, it's easy to have a typo to a synth vocal and not catch it in the mix.

-- David Cuny
My virtual singer development blog

Vocal control, you say. Never heard of it. Is that some kind of ProTools thing?