Tuning and timing vocals in Cubase

What do most people use for tuning and timing vocals and other audio? Cubase Variaudio or third party plugin’s like Melodyne or Autotune? Have anyone use Revoice Pro for vocal production?


Variaudio mostly, Melodyne sometimes.

I use Variaudio because a) it sounds great and b) I love the fact that it’s integrated into Cubase and I can go back and tweak my tuning/timing at any point in the production process.

Variaudio / Audiowarp

Melodyne sounds better but it can be a pain to work with. If you’re working on pop vocals for release however, producers / management / whoever… might insist you work in melodyne.

Wow… I would never pay even the slightest bit of attention to the opinions of management about what pitch correction algorithm I should be using!

Depends who’s paying… I was working with a girl band, and the manager insisted everything went through Melodyne.

Yep, I’m sure that happens. I’ve heard an A&R guy playing a very poorly mixed demo and he said “Don’t worry, they’re going to put it into Pro Tools. Then it will sound great”. I don’t think I’ve ever met a manager or an A&R that would have even have a remote clue about the technical differences between one DAW or another, or between different pitch correction techniques. If a manager says to me “put the vocals into Melodyne”, I would just translate that into “make the vocals sound nice and in tune using whatever is necessary”.

If the guy is literally sitting right next to me with a pay cheque… yeah cool… I guess I’ll put it in Melodyne. Any other day, I’ll Variaudio it. :slight_smile:

Variaudio is often on par with Melodyne sound-wise(for slight pitch corrections, say plus or minus one or 2 semitones), and integrated so easier workflow.
Sometimes variaudio will do a bad job, and Melodyne is fine, so then I use that.
For total restructuring of melodic content, and polyphonic corrections Melodyne rules.
I once analyzed a (fairly simple) 3 minute acoustic guitar part (chords) in Melodyne, exported it to MIDI and there were only a couple of slight mistakes(stray notes with low velociies). We doubled the guitar with an electric piano VSTi, the guitar player was very impressed.

I tend to put vocals into time using cut and slip in the project window. Occasionally I’ll use time-stretch if I need to (using the best sounding algorithm, probably Elastique pitch). I’ve used RevoicePro for aligning backing vocals to the lead vocal in PT and it works very well for this. Never used it in Cubase though.

For pitch changing I generally slap on Autotune as an insert first because I like the way it quickly tightens everything up. If I’m doing pre-production (demos etc) then I’ll use Variaudio quickly, e.g. select-all, quantize pitch 100%, straighten pitch 20%, then leave the Autotune plugin on to help as well.

If I’m doing full production final vocals (for an album) then I’ll generally whiz it into Melodyne (which is a real pain in the backside because we don’t have ARA integration!) and then tune it up carefully, all the time listening to the quality of the vocal shifting. I’ll probably still leave AutoTune on, but I might automate the knobs so that it doesn’t become obvious on any of the notes (again, leaving AT there helps me fix the vocals quickly, i.e. reduces work in Melodyne or VariAudio).

TBH there’s not much difference in sound quality between VariAudio and Melodyne IMHO, they both have trouble with certain vocal sounds, usually when the voice isn’t pure, e.g. a gravelly sounding note or a screaming top note. This is where the artefacts become audible because it’ll get more crunchy sounding. Obviously you get pitch warbling or munchkin effects too sometimes. Actually, this is why I also use Autotune because I find that it doesn’t cause so many crunchy artifacts.

As a final step, particularly for production work, I’ll render the files to new audio and sometimes with particularly tricky bits of ‘emotional’ vocal I’ll cut back in the original untuned vocal using lanes. By this I mean that it’s all about the performance then the quality of the vocal sound, so if I’m unhappy with tuning artefacts then I’ll render it and investigate the possibility of using the untuned vocal to reduce artefacts. Whatever gets the job done really.