Tuning Vocals in Post

[Lunkhead]

...having to open a second DAW just to tune vocals is just a waste of time and energy

That was not at all what I was suggesting. I was suggesting considering spending a little more money and moving to a more fully featured DAW. I haven't used Melodyne lately but I know for certain that the Cubase pitch editing tools offer every feature outlined in that Melodyne YouTube video jan linked. AFAIK the only feature the Melodyne has that other apps don't is the polyphonic auto-tuning support, which isn't relevant for vocals. By the time you've spent $100 for Melodyne you're half way to the cost of Logic Pro X. Spend more on some virtual instruments, etc. eventually you've paid out a higher total cost for your setup. Just something to consider. I know switching DAWs isn't easy or fun.

[grumpymike]

Why ReaPitch and not ReaTune? I'm not sure I understand why the timing gets screwed up either (you mean just if you're moving an item around?)

Oops. That’s what I meant. ReaTune. Yeah, if you place the FX on the item, it uses the timebase of the item and not the track. It’s a lesser known reaper feature, I think. So if you copy or move the item, any time based FX parameters come with it. Whereas if the ReaTune were on the track, you’d have to re-draw the tuning lines.

But it seems you’ve bought Melodyne. Congratulations! It actually works better as a track FX in my experience, so the above is moot.

[grumpymike]

AFAIK the only feature the Melodyne has that other apps don't is the polyphonic auto-tuning support, which isn't relevant for vocals.

Do other products do pitch slope, vibrato, and formant? Serious question. Watch Melodyne’s tutorial videos - they’re kind of amazing. They pick up the harmonic undertones in an electric guitar.

[Lunkhead]

AFAIK the only feature the Melodyne has that other apps don't is the polyphonic auto-tuning support, which isn't relevant for vocals.

Do other products do pitch slope, vibrato, and formant? Serious question. Watch Melodyne’s tutorial videos - they’re kind of amazing. They pick up the harmonic undertones in an electric guitar.

I haven't used Melodyne lately but I know for certain that the Cubase pitch editing tools offer every feature outlined in that Melodyne YouTube video jan linked.

- pitch slope: yes for Cubase
- vibrato: yes for Cubase
- formant: yes for Cubase https://steinberg.help/cubase_pro_artis ... ing_t.html

That video was just vocals I think. I haven't watched any about tuning other instruments. I've used Cubase's to manually tune melodica and monophonic synth to pretty decent effect.

Aw dang, I am just noticing that the Variaudio stuff in Cubase is in the Pro level product only.

[grumpymike]

Oh weird. I would think the formant feature goes with polyphonic capabilities. But what do I even know?

I like how in this video he says “the most common use is to change gender of the vocals”. Well, if “Weird Al” is a gender, then yes, I suppose so.
https://m.youtube.com/watch?v=nPAINeIGxMc

W/r/t Melodyne being a “heavy” plugin, yeah it is, but the vst3 + ARA extension makes it pretty integrated into Reaper. That said, I have run into annoyances with it, which are likely bugs or workflow induced. I wouldn’t mind it being even better integrated, but I also like that Melodyne isn’t held back in their single purpose pursuits. I can switch DAWs at any time and still have the things I paid a lot for. By your logic, we should all just use the stock Garage Band drums, right? :p

[jast]

To answer the question about Essential vs. Assistant: Essential does away with all of the detailed pitch and timing editing. These are the things you can do in Assistant but not in Essential:

• Change the phrasing intensity within notes (from super flat to exaggerated)
• Change the pitch transitions (from instant to gradual - making a pitch transition more gradual can make it sound more natural)
• Fix drifting pitch within a note (or add some)
• Alter formants (only really useful for silly-sounding effects or the occasional note that was sung a tiny bit too brightly or darkly)
• Change a note's amplitude
• Change the timing within a single note (Essential only allows adjusting the overall timing of a note)
• Transient shaping
• Note separation tools (allegedly Essential has a basic version of this and it's more advanced in Assistant? No idea. Having the basic version of this is pretty much indispensable IMO, if Essential has that I don't think this item matters much.)
• Export detected notes as MIDI

I'm not sure on the exact differences in terms of note separation - everything else is useful (some of it very much so) but not absolutely necessary.

[Pigfarmer Jr]

ReaTune. So if you copy or move the item, any time based FX parameters come with it. Whereas if the ReaTune were on the track, you’d have to re-draw the tuning lines.

If I tune with ReaTune and then have to edit (move vocal) I just print/render the track and import the new tuned vocal. But you have to remember to do that. So I was in the habit, for a while, in editing (with tuning or not) and then rendering and importing. I typically only use light compression and a pretty light EQ on our vocal tracks but I'll typically print those as well. I try not to print spacial effects (reverb, delay etc) mainly because I often want to change them later but sometimes because I've sidechained the reverb.

[grumpymike]

ReaTune. So if you copy or move the item, any time based FX parameters come with it. Whereas if the ReaTune were on the track, you’d have to re-draw the tuning lines.

If I tune with ReaTune and then have to edit (move vocal) I just print/render the track and import the new tuned vocal.

You’re describing “the correct way” to do things, just like mastering should be done last, too, and you end up with a leaner project that uses less CPU. But when you’re working rapidly and don’t know your song structure yet, sometimes different workflows work better...

Btw check into Reaper sub projects if you haven’t. They kind of help some of these flows.

[Pigfarmer Jr]

But when you’re working rapidly and don’t know your song structure yet....

For me it is faster to tune, render, and then chop up the track (or tracks) however I like.

Btw check into Reaper sub projects if you haven’t. They kind of help some of these flows.

Never heard of such a thing. Thanks.

[jast]

Semi-related:
So I tried out Pitchmap (which is currently on sale, too) on a few full mixes. I can't say I'd recommend it for that. It misses way too many things for my taste, so if, for example, I try to change the scale, clearly audible parts of the background still use the old scale. It might work for sparser mixes, but I don't think I can justify spending any money on this. I'd rather take my chances with Melodyne's Direct Note Access.

[fluffy]

Oh wow I didn't realize this thread had been revitalized.

I have some thoughts on the "laziness" of using post-recording vocal tweaks (autotune/melodyne/flexpitch/etc).

Once upon a time, when people wanted to play piano, they'd need to tune it. A piano tuner was expected to use a handful of tuning forks to get a few reference pitches, and do the rest of the strings using harmonics and beat frequencies. It took hours to do. Nowadays a piano tuner will just use a microphone and some software that looks like a guitar tuner, and they can get it done much quicker, with better results.

Once upon a time, when you wanted to make music with a synthesizer, you'd have to set up your instrument patches with literal patch cords, and spend a lot of time fiddling with it, and then if you wanted to record multiple notes you'd either need a perfect performance or you'd need to use a step sequencer, which could play 8 notes. To do an entire song you'd have to record 8 notes, then pause the tape, then set up the next 8 notes, and so on.

If you wanted to deal with multiple audio tracks and instruments and so on, you'd be recording them to multiple tracks on tape. If you didn't have enough tracks, you'd need to carefully bounce things down to make room for more things. One slip-up and you lose everything and have to rerecord everything. You also needed an engineer to work the faders as you go, you couldn't just tweak the volume of tracks after the fact, and if you had a 16-track recording that you wanted to bounce out you'd have to adjust the faders manually every single time.

We have modern tools to make things easier. We have softsynths that enable us to get way more sounds than were ever possible on the old synthesizers, able to be saved and and recalled and non-destructively tweaked on the fly. We have automation data for our mixes and effect routing and so on. We have the ability to finely polish every aspect of a recording and a performance to get the sound we want, without needing to spend hours and hours on what amounts to busywork.

And I see vocal performances the same way. We have tools that allow for way more expressiveness than we ever had before, and we can produce much better-sounding and much more interesting music as a result.

We also have ovens for cooking instead of having to build a fire pit, we have running water instead of having to find a stream to bathe in, and as a species we pride ourselves on being tool-makers and tool-users.

When we free ourselves of the constraints of the "proper" (meaning unnecessarily difficult) way to do things, we can do so much more and so much better.

Whenever I see someone complain about the validity of craft based on it not being difficult for difficulty's sake, this is all I can think of anymore:

[Pigfarmer Jr]

We have modern tools to make things easier.

And I see vocal performances the same way. We have tools that allow for way more expressiveness than we ever had before, and we can produce much better-sounding and much more interesting music as a result.

To be fair, autotune is a shortcut for learning an instrument. It's taking a player who may be competent and turning them into something more. It's not the same as building a new amplifier or a new effect pedal. Yes, it can be used that way, and yes I think it's a tool that is more than valid, it SHOULD be used. But is it really the same as taking a medicore guitar player and making them sound masterful?

Interesting is an interesting concept. The problem I see is that sometimes autotune actually makes music less interesting. I like those 70's hits with a little wobble in pitch here or there. The imperfections that make something sound REAL. To be fair, in the 80's those mostly went away for a lot of reasons and can't be blamed on autotune. But if your perfect performance is achieved at the cost of feeling or emotion, is it really better?

None of that means that autotune isn't a useful tool or that it shouldn't be used. My personal feeling is use it or not at your own discretion. But don't be surprised if you continue to hear a few pitchy notes in my performances. I certainly won't shy away from using some tuning (sometimes a lot) but I also won't be shy from not using it, either. The choice is the artists.

[thirdcatmusic]

I won't get into the weeds here; just wanted to add that I use & love Melodyne. I have the bells & whistles Studio version and I rarely use the more advanced features, but they are pretty neat.

[fluffy]

We have modern tools to make things easier.

And I see vocal performances the same way. We have tools that allow for way more expressiveness than we ever had before, and we can produce much better-sounding and much more interesting music as a result.

To be fair, autotune is a shortcut for learning an instrument. It's taking a player who may be competent and turning them into something more. It's not the same as building a new amplifier or a new effect pedal. Yes, it can be used that way, and yes I think it's a tool that is more than valid, it SHOULD be used. But is it really the same as taking a medicore guitar player and making them sound masterful?

But that's not all that it's used for. Obviously artists like Cher and T-Pain have used autotune cranked to the max to provide robot vocals, which is, sure, played out, but I hear so many other bands doing interesting stuff with pitch modification as an effect. The Bird and the Bees' debut album uses heavy use of something (probably Melodyne) to make these amazing lush harmonies. Pitch modification plugins can also do things like manipulating formants to get different tonalities that you wouldn't have available otherwise; I've used this in a few of my tracks to do stuff that my voice isn't physically capable of. It's opened up huge new creative avenues for me (and you probably do not want to hear my raw vocal tracks on those songs except maybe as a cautionary tale of one's reach exceeding their grasp).

Anyway. There's more to vocal pitch modification than just autotune.

Also on the note of a mediocre guitar player sounding masterful, lately I've been using fake guitars in much of my stuff because it's an end-run around my severe chronic pain. So it's either using fake guitars or not using guitars at all for me, a lot of the time. And same goes for, like, modifying my piano takes with quantization and note correction and so on. I've come to the conclusion that it's less important to show off my ability to wank with real instruments (whatever that even means) and more important to make a song that I am happy with.

[jast]

I don't like pitch modification to the extent that natural intonation and modulation is lost. It gets even worse if you use it for harmonies. They always sound super artificial - even though Melodyne tries to add in some humanization, it still sounds robot-y. And, of course, less than subtle pitch changes always add artifacts. I can see that being done in electronica and such, where it can fit the overall vision of the song - but most of the time people just do the same old "iron out the human element" and then it gets super boring to me. If I want perfect pitches throughout I can just use synths and have a lot more flexibility that way.

What's worse is applying pitch correction sloppily or to compensate for a singer singing way outside their abilities. For instance, suppose you have a perfectly fine take except one note is way off, so you fix that note and flatten out the intonation because it was all wobbly... now that note is going to stick out. With e.g. Melodyne's note splitting, pitch transition and pitch modulation/drift tools, you can probably still make it sound natural, but that's more work and many people seem to either not know they can do this, or not care. Bah.

OTOH I do see the merit of using it here and there to fix notes that are off in a way that doesn't have any artistic value... particularly in a production context where you can't just schedule another recording session if you don't have a good take for that phrase. In that sense, it's also perfect for Song Fight where you simply have limited time to go on re-recording forever. I do aim for takes good enough to not need correction (which also means I don't finish songs I wrote that clearly exceed my abilities), but after spending some more time with Melodyne and understanding how to use it properly, I wouldn't categorically dismiss the idea of touching up a few notes for some final polish.

[Pigfarmer Jr]

But that's not all that it's used for. There's more to vocal pitch modification than just autotune.

Agreed.

I've come to the conclusion that it's less important to show off my ability to wank with real instruments (whatever that even means) and more important to make a song that I am happy with.

I hope I didn't imply using these tools was somehow less musical or less artistic. Your examples and a plethora more could be cited to debunk that implication. We should use the tools at our disposal to make the *best music we can. Only the artist can decide what is appropriate. But the audience ultimately decides if they want to listen to it or not.

*best = a myriad of different definitions based on our individual experience

[fluffy]

I don't like pitch modification to the extent that natural intonation and modulation is lost. It gets even worse if you use it for harmonies. They always sound super artificial - even though Melodyne tries to add in some humanization, it still sounds robot-y. And, of course, less than subtle pitch changes always add artifacts. I can see that being done in electronica and such, where it can fit the overall vision of the song - but most of the time people just do the same old "iron out the human element" and then it gets super boring to me. If I want perfect pitches throughout I can just use synths and have a lot more flexibility that way.

What's worse is applying pitch correction sloppily or to compensate for a singer singing way outside their abilities. For instance, suppose you have a perfectly fine take except one note is way off, so you fix that note and flatten out the intonation because it was all wobbly... now that note is going to stick out. With e.g. Melodyne's note splitting, pitch transition and pitch modulation/drift tools, you can probably still make it sound natural, but that's more work and many people seem to either not know they can do this, or not care. Bah.

OTOH I do see the merit of using it here and there to fix notes that are off in a way that doesn't have any artistic value... particularly in a production context where you can't just schedule another recording session if you don't have a good take for that phrase. In that sense, it's also perfect for Song Fight where you simply have limited time to go on re-recording forever. I do aim for takes good enough to not need correction (which also means I don't finish songs I wrote that clearly exceed my abilities), but after spending some more time with Melodyne and understanding how to use it properly, I wouldn't categorically dismiss the idea of touching up a few notes for some final polish.

I mean my point isn't about like, exceeding one's talent, so much as doing new things that possibly couldn't be done by any human. And judging by how this gimmicky track (which predates Flex Pitch and was done purely with automatic pitch correction + formant manipulation) is my most popular song of all time (both on Song Fight! and in terms of sales and streams on all the platforms I'm on) I feel that listeners don't categorically reject things like that either. Of course, it's purposefully sounding robot-y.

Even with natural-sounding stuff, though, I do of course prefer to have a vocal performance that doesn't need tuning, but when it needs a little I have no qualms about doing it, and I'm certainly not going to stop using this stuff for effect in any case.