Jump to content

I can't take Autotune anymore


Mintbeetle

Recommended Posts

  • Members

 

I don't really give a {censored} if it's supposed to be a blatant effect. It makes me want to claw my ears out. The sad thing is, I thought it was cool, or at least not offensive, when it was new. It's just been massively overused.

 

 

A sad fate for many a production technique. It's damn hard to invent something new that ages well.

Link to comment
Share on other sites

  • Replies 96
  • Created
  • Last Reply
  • Members

The problem with auto tune is the is a certain speed of acting above which its blatently obvious and below which you hardly notice it - it aint a bad thing - its just often used badly.

But the again - think plastic voice to go with the plastic tits :p

BTW - I think the effect was originally pioneered for abba - there I beleive they did it by recording flat and speeding the tape up to get the exact pitching.

Link to comment
Share on other sites

  • Members
BTW - I think the effect was originally pioneered for abba - there I beleive they did it by recording flat and speeding the tape up to get the exact pitching.



Naw, it goes back at least to the 1960's. I love Judy Collins, God bless her, but they used tape techniques with her all the time. ;)

Link to comment
Share on other sites

  • Members

I should add that Judy Collins was a lovely singer. That's my point: voice manipulation is used by everyone, from those who can't sing to the finest singers. Judy Collins had a reputation for great singing and so her producers were determined to further that in the studio, no matter what they had to do. It's no different today, and there's nothing wrong with it, if it's a great singer with a couple of flubs and they're off in Hawaii. I'm talking about drawing in a few notes on a recording of talent, not covering up for a bad singer. :)

Link to comment
Share on other sites

  • Members

So far it seems that only Audacity, Cloacal-X, and Birdie really know what Autotune is. It's a simple tool like anything else in the studios toolbox. When used judisciously on an already 'good' or 'great' singer it will be unnoticable. If used to try to 'create' a good or great singer it may be noticable. And of course the 3rd option of using as an effect - well, you WANT it to be noticable. But this can be said of any studio effect. I have 2 regular clients that record aproximately 12 songs each a year. Both of them have nice sounding voices and 'good' pitch. Occasionally, even though we typically do 2-3 takes to create a comp-track, a couple notes need tweaking in a song. In 99% of the cases AT will transparently fix those notes. 1% of the time it will need to be punched (and if you've ever had to punch a few weeks after the initial tracking you know that it can be hard to match the mic placement, pre-settings, comp-settings, and energy of the original).
But saying you hear it GLARING, ALL the time in EVERYTHING is a bit like being a hypocondriac - you want so bad not to hear it that you swear you do. I have had AT haters listen to some of my engineering work and use the same argument - that they can here either it's effect or it's artifacts in the song. By simply pulling up the raw tracks I can prove that it wasn't used where they swear they hear it.
Of course you are right in that many current radio hits use it ad-nauseum, but I believe that's a simple case of trying to create (or better said, conform) to a supposed succesfull methodology - much like using a MiniMoog during the 70's wether the song called for it or not.
But AT is simply another tool that if you actually learn how to use it correctly, can really polish up a vocal arrangment. And I believe that an engineers reason for using it is a lot different than the producers reasons. And I can't really sign-off without saying that that T-Pain clip was arguably CRAP anyway.

Oh yeah, almost forgot - to -ANTI- try this station for a few days, it's like ANTIradio:
http://minnesota.publicradio.org/radio/services/the_current/songs_played/

Link to comment
Share on other sites

  • Members
So..."Praise You" by Fat Boy Slim......that would be a vocoder?


As said, that's just finding a loop point and looping sample to create that effect. But you're on the right track. Autotune can add that sort of effect to a sustained note. Your ear picks up that its almost too pitch perfect and something sounds wrong. That's the main way you can notice it on tracks where you're not supposed to notice it, as well as any kind of rapid pitch change which can sometimes create a little bit of a "warble" effect as the software tries to track it. Usually you only notice that warble type of noise when it's being used on large parts of the performance, and not just to correct a note or two. As already said, when it's done for just a note or two it's unlikely you'll notice. I'm sure, thanks to a sustained note sounding "too perfect" sometimes people think, wrongly, that autotune is being used, when it's just that the singer hit that note more exact than is usual :)

And it can be used live, and probably is by some people.

Link to comment
Share on other sites

  • Members

 

So far it seems that only Audacity, Cloacal-X, and Birdie really know what Autotune is. It's a simple tool like anything else in the studios toolbox. When used judisciously on an already 'good' or 'great' singer it will be unnoticable. If used to try to 'create' a good or great singer it may be noticable. And of course the 3rd option of using as an effect - well, you WANT it to be noticable. But this can be said of any studio effect. I have 2 regular clients that record aproximately 12 songs each a year. Both of them have nice sounding voices and 'good' pitch. Occasionally, even though we typically do 2-3 takes to create a comp-track, a couple notes need tweaking in a song. In 99% of the cases AT will transparently fix those notes. 1% of the time it will need to be punched (and if you've ever had to punch a few weeks after the initial tracking you know that it can be hard to match the mic placement, pre-settings, comp-settings, and energy of the original).

But saying you hear it GLARING, ALL the time in EVERYTHING is a bit like being a hypocondriac - you want so bad not to hear it that you swear you do. I have had AT haters listen to some of my engineering work and use the same argument - that they can here either it's effect or it's artifacts in the song. By simply pulling up the raw tracks I can prove that it wasn't used where they swear they hear it.

Of course you are right in that many current radio hits use it ad-nauseum, but I believe that's a simple case of trying to create (or better said, conform) to a supposed succesfull methodology - much like using a MiniMoog during the 70's wether the song called for it or not.

But AT is simply another tool that if you actually learn how to use it correctly, can really polish up a vocal arrangment.

Absolutely.

 

I have dozens, if not a hundred or more tracks that have been Autotuned correctly. Yes, you can kinda tell it working on one or two singers, but that's because they sucked to begin with. I can post a few tracks and will give $100 to the person who correctly picks an Autotuned note.

Link to comment
Share on other sites

  • Members
And furthermore, to extend the argument, do we really only want bands that are great live? Don't we also want studio bands, bands that only excel in the studio? I certainly couldn't live without my Beach Boys recordings. Or Cocteau Twins. Or . . . name your poison.
:)



That's what I'm saying. If I see an artist/group live I expect them to be able to play/sing. If I buy a record I want it to sound good, and I downright expect some studio trickery to be employed in the process. "We want our film to be beautiful not realistic".

Link to comment
Share on other sites

  • Members
So far it seems that only Audacity, Cloacal-X, and Birdie really know what Autotune is. It's a simple tool like anything else in the studios toolbox. When used judisciously on an already 'good' or 'great' singer it will be unnoticable. If used to try to 'create' a good or great singer it may be noticable. And of course the 3rd option of using as an effect - well, you WANT it to be noticable.



If you correct only the off notes in editing, that is the correct way to use auto-tune. And you are right, it won't be noticeable.

The problem comes when auto-tune is applied "naturally" to the entire song, whether the singer justifies it or not. For mediocre singers, it does make them sound in tune, so it's justified there for better or for worse. For good singers, fully autotuning a song tends to eliminate some of the nuances of the voice, so it's not justified IMHO. As noted, vocal manipulation is nothing new in the music business, but autotune may encourage laziness.

I don't think autotune is 100% present in pop music, either. I think you get some passion about it, though, because auto-tune is perceived as a cheat, and you do hear it pretty obviously with some artists who, despite being mediocre singers, do well in the pop music machine.

Of course you are right in that many current radio hits use it ad-nauseum, but I believe that's a simple case of trying to create (or better said, conform) to a supposed succesfull methodology - much like using a MiniMoog during the 70's wether the song called for it or not.



That's pretty much it. Same deal applies to brick-wall compression, which as a paradigm I'm surprised has lasted as long as it has. Maybe some day, autotune will be seen as "retro" like other paradigms of the past -- remember when every snare drum had to have a boomy echo? :)

Link to comment
Share on other sites

  • Members
Nope. There will be some songs, generally the ones with good singers to begin with and good engineers working with them, that go right past your ears, with subtle and discreet use of autotune. You should say, when a song is blatantly autotuned, you hear it. Sorry, that is the truthy truth.


In the old days, even the finest artists would blow a note or two, leave the studio to go on vacation, and then the poor producers and engineers would spend days speeding up/slowing down reels very slightly, and syncing and comping them. Autotune simply eliminates that, if done right. I wish I had it on one album I recorded, back in 1989. My solution? It was only the opening note that was off, so . . . we faded it in. If that happened today, it would be autotuned and you would never know it, I guarantee it.


And yes, I have very, very good ears. It's my best feature.
:)



Then you hear them live, without graphic mode, and it's apparent whether they're actually good or not. Sorry, but you can't really fudge that one.

It's getting so sad though. I'm not much of singer, but if I can't get a take well, I try it again till I get it right (or move and accept that it's out of my range). I have the tools, but I don't want to use them on myself.

Melodyne says I'm sharp on most notes. I don't really give a {censored}. I'm not sharp, according to my ears, which happen to listen better than a machine. It's completely wrong. Why should I trust it to automatically get things right? It usually won't.

Link to comment
Share on other sites

  • Members

this is then where your problem with automatic pitch correction lies...
the automatic part.

autotune does have a manual mode and if used by someone with decent hearing and an understanding of pitch and natural performance sound it can be invisible.

in automatic mode it just plain sucks. the detection isn't that bad but the assumption is to eleminate too much pitch variation from the pefect note. the basic assumption of perfect pitch is where automatic mode fails. there is always going to be some deviation from perfect from instance to instance in even the most perfect singers. a note in the vocal slides from one pitch to the next and the slide rate is what we classify as a more or less perfect vocal performance. the issue is that in automatic mode that variation of slide and natural pitch tracking of the singers ear to slide again to correct pitch is what autotune destroys.

when a singer sings a sour note you can tell it's off by ear. if a singer is sliding the pitch up a bit late to adjust into correct pitch that is natural.
automatic mode doesn't differentiate. it's off by such a percentage it gets shifted to perfect period.

you do it by hand on only those parts that are flubbed and you get something that works.

Link to comment
Share on other sites

  • Members

Melodyne says I'm sharp on most notes. I don't really give a {censored}. I'm
not
sharp, according to
my ears
, which happen to listen better than a
machine
. It's completely wrong.

 

That's the right attitude there ... trust your ears, not the engineer who wrote the algorithms inside box ...

 

There's no correct pitch, there are just ones that sound right ... and this sounding right business is very culturally etc dependent matter. One thing DX7 brought with it is this what I call 'Yamaha engineer tuning' ie mathematically correct equal temperament ... which sounds really off for most purposes except if you want to sound like a robot. Pianos btw are never tuned to exact equal temperament -- even if wikipedia says so. :)

 

That Oasis singer was consistently flat ... too bad they didnt have Autotune to do those vocals, Oasis could've made it big then! :p

Link to comment
Share on other sites

  • Members
Melodyne says I'm sharp on most notes. I don't really give a {censored}. I'm
not
sharp, according to
my ears
, which happen to listen better than a
machine
.


They don't :D.

But a machine will listen at the well-tempered frequencies. You know what happens when 3 people each sing a note of a choir? They tune up or down in very subtle ways so there's no "beating" effect (try this yourself; it's what gives the 10cc "mixing desk tape choir" its charm), and this is what Hermode tuning actually does on the Virus and Q.

It's a tendency to do that; you can't really avoid it unless those 3 people are in separate rooms. You hear a chord in the background and tune your voice in such a way that you minimize the beating; or at least, that's the only explanation I can think of.

It's completely wrong. Why should I trust it to automatically get things right? It usually won't.


The difference between you and the machine is that you can break the rules and the machine doesn't, because 1) it can't figure out the context (look how hard it was even to extract a single note from someone singing solo; let alone a complete track) and 2) it doesn't know when to break those rules.

Link to comment
Share on other sites

  • Members

I just want to thank everyone here for this great topic.

For the first time, after many years, this is the first thread I have read that didn't evolve into the great debate about how Cher recorded "Believe" and that SOS article where the producers took the story writer for a real "ride" and got a good laugh out of it. This to me is a new milestone.

And when used properly, you cant hear voice correction whether it is Autotune or Melodyne.

The problem is, just like hundreds of other software tools especially limiters, they are being abused and used incorrectly. (unless you are after the obvious effect such as Cher, Faith Hill, etc)

Link to comment
Share on other sites

  • Members

Interestingly (or not) enough, we'll occasionally monitor certain singers through Autotune in auto mode (but track dry and they monitor themselves dry), because after a long session, our ears start playing tricks on us, and it can become difficult to tell how bad the pitch is. We'll end up listening to Autotune artifacts instead of the intonation

Link to comment
Share on other sites

  • Members

That's what I'm saying. If I see an artist/group live I expect them to be able to play/sing. If I buy a record I want it to sound good, and I downright expect some studio trickery to be employed in the process. "We want our film to be beautiful not realistic".

 

 

Hmm - I guess it depends what you go for - I expect a good show with good energy, good atmosphere - how they acheive it - thats upto them. As for auto-tune - well there is a hardware box that includes it in the form of the AVP-1. Alot of bands use them for one reason or other live - the box does alot more then just re-pitching.

Link to comment
Share on other sites

  • Members

They don't
:D
.


But a machine will listen at the well-tempered frequencies. You know what happens when 3 people each sing a note of a choir? They tune up or down in very subtle ways so there's no "beating" effect (try this yourself; it's what gives the 10cc "mixing desk tape choir" its charm), and this is what Hermode tuning actually does on the Virus and Q.


It's a tendency to do that; you can't really avoid it unless those 3 people are in separate rooms. You hear a chord in the background and tune your voice in such a way that you minimize the beating; or at least, that's the only explanation I can think of.



The difference between you and the machine is that you can break the rules and the machine doesn't, because 1) it can't figure out the context (look how hard it was even to extract a single note from someone singing solo; let alone a complete track) and 2) it doesn't know when to break those rules.

 

Yes. And that's why it's wrong. What happened when there wasn't autotune? Did somebody strike keys or have perfect pitch? Is that too accurate or not accurate enough?

 

Seems to me it was plenty accurate. It's this whole mindset of trusting the machine for more than it was meant to do when all it gives you is its "result" of what you should be singing overlaid onto what you were singing. Even with no correction, most tuning programs introduce artifacting right off the bat. With Autotune it's usually phase, with Melodyne it's phase, granulization, etc. So even if I don't do anything with them, they're already hurting my recording a little bit, and no engineer can erase that. It's just overtones no one will really hear...

 

Kind of like in radio... How much distortion can they induce within the acceptable range? It's always there. You can't get away from it.

 

EDIT: And if you think ears don't listen better than a machine, you might not wanna try your hand at engineering.

Link to comment
Share on other sites

  • Members

 

Hmm - I guess it depends what you go for - I expect a good show with good energy, good atmosphere - how they acheive it - thats upto them. As for auto-tune - well there is a hardware box that includes it in the form of the AVP-1. Alot of bands use them for one reason or other live - the box does alot more then just re-pitching.

 

 

Well, that's true enough, the main thing is how engaging the performance is. However, I do think that a live show should feel like something special, and if I think the people on stage are hacks who are just miming along or otherwise not genuinely presenting themselves in the performance, that seriously detracts from my enjoyment of the show.

 

But I do agree that even musicians with humble abilities can put on a great show if they present themselves in earnest.

Link to comment
Share on other sites

  • 2 weeks later...

Archived

This topic is now archived and is closed to further replies.


×
×
  • Create New...