The singers don’t require it, the listeners do. Back when pitch correction wasn’t a thing, a certain amount out of tune singing was expected, so nobody complained. Once autotune became a thing and people heard perfect pitch on songs, they started to demand it. All those professional singers who didn’t need it 30 years ago would need it today if they wanted to get anybody’s attention.
The fact that you are describing the notes as ‘pitchy’ means they need fixing. In answer to your query I’d bet that close to 100% of commercal recordings have some sort of pitch correction.
I can’t think of any time I ever heard a flat note that didn’t detract from a song. Flat notes make me nervous.
I was asking, how many pro singers lack the ability to sing their own songs with correct pitch? I understand that there are some pretty technically mediocre, commercially successful singers, and they would definitely benefit from having their vocals auto-tuned, but there’s probably many pros that don’t require auto-tune except in the occasional instance. In the case of an occasional flub, the pitch correction would be useful and fitting to use if the singer can’t or doesn’t want to take a moment to rerecord the phrase again.
So yes, nothing wrong with fixing a bad note or notes. I’m just not convinced that every tiny, inaudible micro-cent, out-of-tune vocal needs to be corrected.
There are many, many records that date back to the 80’s that would gain no improvement from auto-tuned vocals, even by today’s standards. I mentioned Ozzy (Sabbath) and Robert Plant (Zeppelin) having some pitchy moments on their records but there were many vocalists that were very precise with their pitch…even some excellent vocalists in the 70’s. By the time the 80’s rolled around, productions and audio precision was very good in large part.
It is often assumed that pitch correction is synonymous with the late 90s/2000s after the advent of Autotune and similar inventions. However, tuning vocals started even in the days of tape in the 70s, and certainly in the early 80s…
Bob Clearmountain and others have recounted using early digital samplers to correct singers’ pitching in first half of the 80s. - “flying” it off tape into a sampler, then back onto tape after correction. It was much more difficult and time consuming to do then, and much less common than it is today, but it was certainly done.
1 Like
OK so now we need to define terms. Originally you said “how shitty of a vocalist do you gotta be if you need to use autotune!!!” Now you seem to be saying that if only certain specific notes are out of tune and are corrected, it doesn’t count as ‘autotune’.
Autotune does not correct notes that do not need correction, it only corrects the ‘pitchy’ notes. So whether the pitch correction is done automatically or manually is a matter of semantics in my view. Pitch correction is pitch correction however it’s dressed up. So if we’re saying that it doesn’t really count as ‘autotune’ if only a certain numer of notes need correction, the question is, how many notes would that be? Or maybe we should represent it as a perentage of the overall total number of notes sung?
Or maybe we could simply say that if it needs any pitch correction at all, then the singer ‘requires autotune’?
Your view that pitch correction would not improve early Ozzy and Plant recordings is subjective. Led Zeppelin is my favourite band of all time, but countless times I have found myself wanting to hear certain notes sung by Plant corrected.
Ozzy, although a trailblazer who has rightfully cemented his place in rock and roll history, has never been a good singer (in my subjective view 
). Virtually every track he ever recorded in the early days would benefit from pitch correction in my opinion.
Again, at first you say “it’s easy enough to sing in key”, then later you say it doesn’t matter if a few notes are out of tune. Well if - after giving it your best shot - a few notes are out of tune, then it can’t be that easy to sing in key.
In my experience of recording hundreds of singers, I believe that that no-one can sing a song perfectly in tune from start to finish - even the best singers in the world.
So if the best singer in the world records a lead vocal and the producer has it in his/her power to make it even better, what is going to happen? He/she is going to make it even better.
I don’t see pitch correction as being “better” most of the time. It’s just the standard today. It’s what people expect to hear. It’s what professional recordings sound like.
Part of what gave each era its own sound was that standards by which they recorded. Today, pitch correction is a standard. It’s what people expect to hear. Pitch correction in the 70s wouldn’t have sounded like the 70’s, it would have sounded like a weird hybrid. Not better or worse, just a different sound.
I would argue that you can’t get the modern vocal sound without pitch correction, because pitch correction is a key ingredient in the modern vocal processing chain. For the same reason you can’t really have a modern sound without compression.
It’s not better vs worse or a skilled vs unskilled or time vs no time thing. It’s a standard. It’s what people expect and demand to hear. There are plenty of valid arguments for thinking it’s an annoying standard, but it’s a standard nonetheless.
3 Likes
Well stated, Boz!
No, it’s obviously still auto-tune.
I’d be interested to hear certain parts of those old songs auto-tuned to hear the difference. I’m don’t know why these old bands and their handlers wouldn’t do a re-release of their music with auto-tuned vocals. We’ve seen many remastered albums from those 70’s and 60’s era bands. Maybe the vocals have been auto-tuned on some of those remixed/ remastered albums?? I have no idea.
It is easy enough to sing in key/ in tune if you’re singing within your vocal range and you’re a capable vocalist. Of course no singer is perfect, just as no guitarist, bass player or other instrumentalist is perfect. If a vocal deviates by a mere cents (micro) that’s not a problem…Almost all instruments deviate on a micro level. So, I’m not talking about robotic/ computerized perfect singing.
I think there are many good singers who can. Again, I’m not talking about micro deviations that the human ear can’t detect. I’m talking about obvious and clear out-of-tune note deviations.
I just had another thought that’s not specifically about what we’re talking about: …
It’s interesting that auto-tune is used to perfect imperfect vocal performances by humans but with drum machines/ samples/ virtual instruments we use the humanize function to make the drum performance imperfect.
1 Like
I love Ozzy’s tone but he has really bad vocal control/ technique…But he doesn’t seem to have very good motor control of his other body movements either. 
I wouldn’t doubt that. Do you know if pitch correction was done on Def Leppard Pyromania or their Hysteria album?
We have a difference of opinion. Singing is the singlemost difficult discipline to deliver in a studio. You either have lower standards of acceptability than others, or your discernment is impaired. (I don’t mean that as an insult. The ability to discern off-pitch notes varies among us).
Show me a good singer and I’ll show you someone whose performance can be improved with Melodyne.
Everybody talks about “back in the day” - they did pitch correction before melodyne, they would speed up or slow down tape to pitch the note. What is the difference in that vs to composer who places midi notes on a drum beat, a synth track, a piano track? I am not the best piano player as guitar is my main thing. But I can’t always play the composition I hear in my head, so I will place midi notes as necessary to complete a composition. Doesn’t make you less a musician, it is part of the music making process. Just like using drum samples (which was also done “back in the day” as well). How many bands you know and love, didn’t play on their own album, studio guys did.
This is true but there is more to the story. The subject of proper intonation has been debated for centuries, the main problem being the difference between ‘just intonation’ and ‘tempered intonation’. Here is a quote from an article on the Sound on Sound website.
“It seems ironic that we often use pitch–correction tools to apply equal temperament to sources which don’t need or suit it. Singers who are harmonising an existing melody, for instance, will typically aim to do so in the purest possible way — so it’s quite possible that if we blindly and automatically run their contributions through a pitch–correction plug–in, we might actually be making them more out of tune, not less; and it may also be that the characteristic ‘sound’ of vocal pitch-correction results partly from forcing voices to intervals they wouldn’t naturally use.”
Here’s the full article, you’ll find this passage towards the bottom of the page.
These are the key words. Personally I don’t do anything blindly and automatically. I listen, then I correct when and where - in my judgement - it is necessary.
I’ll add to that that I used to use Melodyne in a pretty dumb way. Select all -> correct pitch (and maybe pitch drift too.) Done. That sometimes resulted in my vocals being more out of tune than they were originally, because of the way the software grouped the notes. You have to be really careful where you slice and tune them. I used to use Melodyne as a stand-alone software, but now I only use it as a VST in my DAW so I can listen back to the vocals with the track as I tune them.
Wicked, here’s my take for what it’s worth. Back in the classic rock and 80s metal days, budgets were enormous. Now they are way smaller. There was a time where you could sing it until you got it as right as it was going to be for a particular singer. Then one day things changed.
Comp tracks were created. Meaning, a singer would sing the part 5-10 times and an audio editor and producer would create a vocal track from all the best parts. The problem there? It was still time consuming and costing money.
Antares came out and made a splash with T-Pain and especially Cher due to them over using it as an effect. It was used subtly with other artists but the artifacts were still huge even when used sparingly.
Melodyne with DNA enters the picture and now we have a seamless pitch correction product that will only show forth artifacts if you want it to or don’t know how to use it. You can literally play one chord on a guitar and create others from that one chord that sound so real, no one but you would know you literally created all the other chords in the passage.
This in turn streamlined vocal correction and it not only made it more believable, it was way faster and saved a boat load of money and editing time. Someone has to really sing something horribly wrong to where Melodyne can’t save them. I’ve actually had that happen a few times. Lol!
Also singing in key may be easy for YOU. But to others, it is the most challenging thing they’ve ever had to do. Take me for example. I’m somewhere in between relative pitch and perfect pitch. Meaning, sometimes I hear a note and know exactly what it is and can call it out, other times I fail but come close.
I sing in key about 98% of the time with very little drift/human error. On my new album, this one song just sounded funny to me. No matter how hard I tried, I just felt the vocals sounded funny. I fired up Melodyne and sure enough, I was almost a half step sharp. I had never done anything like that before. I fixed the vocals via Melodyne but it really bothered me because nowhere else on my album did I use Melodyne like this. Here or there on a word or whatever, but never an entire vocal track.
So I looked into it deeper and rerecorded the vocals but this time cut the bass guitar out. Checked with Melodyne, still sharp. Rerecorded with just bass, bam…vocals were as near perfect as humanly possible for me. My rhythm guitars had somehow increased in pitch in between correct and half step up. Singing with them and keying off them was the problem. I didn’t think to Melodyne them, so I just rerecorded them, sang the vocals again and everything was just right.
This above happenes to quite a few. I’m a tuning stickler so I’m a bit surprised as well as embarrassed that this got by me. But, I’m only human and stuff like this is bound to happen when you record, perform, mix, master and produce your own album with all the stress of a few labels and other industry people at your back.
But that said, today it’s all about working smarter not harder. Most of the singers of today are using Antares live because they simply can’t sing. I never want to be something I’m not. When I’ve fixed things with Melodyne, there were reasons. Some of my vocal sessions went on for 7 hours. I simply had no voice left and no time to re-sing certain things like stacks of back up vocals. You don’t want them to sound like one man anyway. That was my other reason for using Melodyne. I could use the formant control to alter the voices so they don’t sound like just me singing the back up parts.
But on lead vocals, I’ll sing any of my new songs in real time and nail every part with no correction. The idea is to use it as a tool/weapon, never a crutch or a means of survival. 
2 Likes
Someone gave me a reference for this once, he said within 18 cents is pretty good vocal pitch or tuning. I’m not sure why that number but have used it as a reference since then. It may be a bit liberal as it’s probably more variance than a well-tuned guitar which might vary +/- 5 cents? But the vocal instrument is a tricky one so it may be a good standard as an occasional maximum. Any more than that would probably get a “Yo dog, it’s pitchy” from Randy Jackson. 
1 Like