For years, every vocal I tuned or aligned in Live ended up printed. Melodyne or VocAlign sat on the track, I played the part in, did the edit, and then froze and flattened the track or recorded it to a new one so I could close the plug-in. Whatever was on the track after it went into the audio too: the EQ, the compressor, the de-esser, sometimes the saturation.
It worked. But it meant I made mix decisions at the moment I was least ready to make them.
The vocal chain isn't finished when the tuning is
When I tune a vocal, the song usually isn't done. The beat still changes, the arrangement gets thinner or fuller, and the vocal chain moves with it. A de-esser set against a sparse verse is wrong once the hi-hats come in. A compressor that sounded right on the rough mix is too much on the final one.
Once the chain is printed into the audio, every one of those changes means going back: unfreeze if I still can, change the setting, print again. Or, more honestly, I didn't go back. I left the compression a bit too heavy because reprinting four doubles wasn't worth it that evening.
Printing twice is how vocals get worse
The worse version of this is printing on top of a print. Tune the lead, print it with the chain, then later add another compressor to the track because it needs more control. Now the vocal has gone through two compressors, one of which you can't change anymore. Do that for a few rounds of revisions and the vocal gets smaller and harder without anyone deciding it should.
Doubles make it worse. If the lead is printed through its chain and the doubles through theirs, at different times, against different versions of the mix, they stop sounding like one performance.
What I do now
I keep three things apart until the end:
- The performance: the audio itself. Tuning, timing and comping change this, and nothing else.
- The sound: the vocal chain on the track. EQ, compression, de-essing and saturation stay plug-ins, so I can change them up to the last bounce.
- The space: reverb and delay on returns, as they always were.
A tuning fix changes only the performance. A new compressor setting changes only the sound. Neither forces me to redo the other.
This is also why I built EDIT and ALIGN the way they work. With EDIT, the EDIT Bridge sits first on the track, so I hear the edit through the whole chain while I tune, and Apply writes back only the edited audio. With ALIGN, the aligned Dubs go back on their own tracks and their chains stay where they are. In both, the original clip stays untouched until I press Apply, and Revert gets it back.
When printing is still right
I still print. Just later, and on purpose:
- To finish a song. When the mix is done, printing the vocal with its chain is a sensible way to close it.
- To hand it over. A mixing engineer or a collaborator without your plug-ins needs audio. Send them the dry tuned vocal and a printed version with your chain, and let them choose.
- For sound design. Resampling a vocal through something wild and chopping the result is a creative move, not a cleanup step. Print as much as you like.
- To save CPU. If a big Set struggles, freezing a finished vocal is fine. Just keep the unfrozen version around.
Try it on your next song
You don't need any of my tools to work this way. Tune or align on a clean track with nothing after the editor, print only the edited vocal, and move it back to the track with your chain. It's an extra step every time, but you get the separation.
If you'd rather skip that step, EDIT and ALIGN do it for you. The guides on tuning a verse and its doubles and aligning backing vocals and harmonies show how it looks in practice.
