What Key do you put your vocals in when using Auto Tune?

Absolutely the key of the song for me for rubbish singers, but for better ones, I set it to chromatic - then they can sing the blue notes if they want.
 
I am in the Rob camp. Chromatic, but not a 100% as I want the vocals to be close to the pitch of a note, but not spot on. Sometimes off notes work for the song.
 
I don’t use auto tune. Not worth it for me. I’m pretty good at singing in key. I understand even the pros use it now, and I wish they didn’t. I can hear it.
 
I've never used the "auto" part of Melodyne (what I have), and it's very rare that I use it at all. But if I do, I set it to the key of the song so I can see where the notes are landing. If I hear something that sounds pretty far out I can see what is actually happening and make a manual adjustment. Sometimes it's just removing some of the "warble" or maybe it's splitting something and moving just a part, or lowering gain at a start or finish that drifts. Whatever makes it stop sticking out.

Not saying I couldn't use it more, but maybe OP is talking about those songs where the entire vocal track has that [weird] robotic sound....
 
I've never used the "auto" part of Melodyne (what I have), and it's very rare that I use it at all. But if I do, I set it to the key of the song so I can see where the notes are landing. If I hear something that sounds pretty far out I can see what is actually happening and make a manual adjustment. Sometimes it's just removing some of the "warble" or maybe it's splitting something and moving just a part, or lowering gain at a start or finish that drifts. Whatever makes it stop sticking out.

Not saying I couldn't use it more, but maybe OP is talking about those songs where the entire vocal track has that [weird] robotic sound....

That’s what I assumed he was talking about. I hate that sound.
 
I must admit I treated autotune, and I have had an Antares rack unit for years - as a problem fixer, until the Cubase update where you could take an audio file, and cubase would graphically show you the pitch of every note, and let you change them, pretty seamlessly too.

Even Opera singers - proper trained ones occasionally slip, and comping opera is now possible and nobody need know. One of the big snags with these singers is that they cannot repeat and repeat, and dropping in type recording robs them of the feel - and they need to match the volume and tone. They get worse when you have to say "can we do one more?" Sitting with a trained singer, they can be mega critical. They spot cents and once they know they can be fixed, they want perfection. It's very strange. When I introduced the technique, I expected rebuttal - but no, their goal is perfection, so I got quite used to getting emails - in bar 22, when I sing 'celsis', the 'ce' is a tiny bit flat, can you fix it please. I think it even generates a little more money. Budgets are always tight, and my preferred way of working is to sit with them in the studio while you do a few edits. They can see how long it takes to get it right - and then when they are sitting at home with their feet up and hear a small slip, they know it will take me maybe half an hour to fix it, and re-dave everything, and they know the costs - but to them it really is worth it. When you look at the syllable in question, you find they are right. When I'm doing middle of the road stuff, some singers slide into the notes stylistically - but often, they over-cook it, and they want that fixed.

I've just put out a piece by a singer who wanted to try three or four songs, intending to pick just one, and I've discovered I actually had enough takes to complete the entire song, from what were essentially outtakes. After getting the completed vocal track edited into one, I went back and fixed every wrong note. Some were so wrong because she didn't really know the melody properly, so sung in tune, the wrong notes. I've fixed it, and she's very happy with it. A song from nothing!

Autotune in the traditional sense is probably now too blunt a tool, but the software versions can be so much more flexible.
 
I don't have Melodyne or Antares, just the one built into Reaper. I've used it a bit on some tracks, but its not to get the robot voice. I've with Crows on that, I hate hearingit, especially when it's all over the vocal through the whole song. If it was used for an effect during some portion of a song, ok. Otherwise it would be like using a wahwah pedal on all the guitar parts of a song.

I don't use it on auto mode. There were a few notes that I manually adjusted some notes that just sounded off.
 
I've never used it and I don't use it for vocals.
I use threats of apocalypse and mayhem. :eek:
Seems to work well. :D
 
A sampler is a much better vocalizer than an autotuner. Once the vocal is down , normalized and stripped, it is graphed to a master BPM. Once encoded you can lengthen words , hold notes, play the melody on the keyboard. choose the specific notes to hit. Play the whole damn choir. Play chords for harmonies. Roland and AKAI products from Japan were excellent vocal designers.

Check out the Roland V sythns or AKai has a similar one

Set it to chromatic.

Watch the gregorian chant at 10 seconds to 20 seconds. Vocals are samples. Samples are vocals.
 
Sorry, but that voice is just as robotic as the Tpain or any other Autotuned voice. Not my cup of tea unless you're shooting for an effect.

I can't imagine CSN doing Suite Judy Blue Eyes through that thing.
 
I must admit I treated autotune, and I have had an Antares rack unit for years - as a problem fixer, until the Cubase update where you could take an audio file, and cubase would graphically show you the pitch of every note, and let you change them, pretty seamlessly too.

Even Opera singers - proper trained ones occasionally slip, and comping opera is now possible and nobody need know. One of the big snags with these singers is that they cannot repeat and repeat, and dropping in type recording robs them of the feel - and they need to match the volume and tone. They get worse when you have to say "can we do one more?" Sitting with a trained singer, they can be mega critical. They spot cents and once they know they can be fixed, they want perfection. It's very strange. When I introduced the technique, I expected rebuttal - but no, their goal is perfection, so I got quite used to getting emails - in bar 22, when I sing 'celsis', the 'ce' is a tiny bit flat, can you fix it please. I think it even generates a little more money. Budgets are always tight, and my preferred way of working is to sit with them in the studio while you do a few edits. They can see how long it takes to get it right - and then when they are sitting at home with their feet up and hear a small slip, they know it will take me maybe half an hour to fix it, and re-dave everything, and they know the costs - but to them it really is worth it. When you look at the syllable in question, you find they are right. When I'm doing middle of the road stuff, some singers slide into the notes stylistically - but often, they over-cook it, and they want that fixed.

I've just put out a piece by a singer who wanted to try three or four songs, intending to pick just one, and I've discovered I actually had enough takes to complete the entire song, from what were essentially outtakes. After getting the completed vocal track edited into one, I went back and fixed every wrong note. Some were so wrong because she didn't really know the melody properly, so sung in tune, the wrong notes. I've fixed it, and she's very happy with it. A song from nothing!

Autotune in the traditional sense is probably now too blunt a tool, but the software versions can be so much more flexible.

You pretty much summed up pitch correction use for me. I find it better to capture the emotion of a performance first, rather than hashing it out with multiple takes to pitch perfect. There are no million dollar budgets for members recording in my studio. And not everyone is Geoff Tate or Pavarotti either...

I have never set anything to any scale. Will never use Melodyne or any pitch control software to mimick that 'T-Pain' thing. I find that to be the most annoying thing in music ever. But that is just my opinion. I choose not to because it makes my brain crazy...

YMMV. Ha! I also swore I would never type that either so... lol
 
This is a auto tuner thread. I was suggesting that a sampler allows greater control over the source material. The demo videos should reflect that.

During the chorus these faked vocals are a good reinforcement, until you have back up singers.

Not for CSN. Perhaps M83.



A civilization that does not appreciate the beauty of M83 will not make it to 3000.
 
Beak, I know you love your Roland, but any way you cut it, it makes a vocal sound processed and altered. Its a gimmick, which is perfectly fine if you want that effect. For me it's no different that what King Crimson did to Greg Lake's voice on 21st Century Schizoid Man. But that same touch wouldn't have worked for I Talk To The Wind.

It's not good for pulling a real voice sound back into tune without destroying its natural timbre.

As for M83, I've never heard of them. That's not surprising. There are simply too many groups out there to know them all. I have a buddy who feels about the same about Evanescence.
 
Back
Top