How to get that "Sweet" audio

A

Ashbury

Active member
Hi guy's. I was just wondering if there were certain techniques or ways in post to have your audio sound like any of the key and peele sketches. Or for instance in the new Revenant Doc at about 20:30 mins in with the women talking https://vimeo.com/157784484

I would love to hear some suggestions or how I could pull this off!
 
OK...this is simply PRO audio.

What you're hearing is CLOSE, clean, proper audio using a VERY GOOD mic and mic preamplifiers..

The fullness and smoothness comes from being close to your subject with a very good mic.

That's really pretty much all there is to it.

Although POST processing will push it even further.

Very careful and moderate peak limiting smooths out volume spikes, and careful EQ'ing GENTLY can add a wee bit more fullness and often lessen harsh frequencies.

But being a good recordist is like being a good photographer.
Being a good sound mixer is like being a good photographer.
You do have to learn the craft. You don't just buy a good camera and suddenly all your pictures are great.

But of the 2 other disciplines, I would say that capturing audio is easier then putting a good mix on it. If you simply follow the rules of putting a good mic in the right position with correct gain-staging on your audio devices then the audio you capture will be about as good as it gets.
MIX, however.... more people make things WORSE rather than better when they start mixing. THAT darn skill takes a lot of practice.
 
Last edited:
Ashbury said:
I would love to hear some suggestions or how I could pull this off!
Click to expand...

As DPStewart says, that's the sound of a professional boom op and mixer. Putting a excellent mic into proper position, running the output to a excellent field mixer, and getting rid of extraneous noise during taping. They make it look so easy and effortless, just like Fred Astaire and his dancing. Just like Stephen Curry shooting 3s. Takes talent and lots of practice. With an emphasis on the "lots of practice".

Once you have clean professional dialog capture, you can polish it a bit in post. Some compression, some EQ, etc. You can even do things like pitch shifting if desired. But it all starts with a solid clean capture; you can't reverse engineer it -- you can't create a clean capture in post.
 
Bruce Watson said:
They make it look so easy and effortless, just like Fred Astaire and his dancing.
Click to expand...

Of course on a sound related note the sound of his dancing was done in post on a Foley stage with his choreographer doing the tap sounds.

One of the things that drives a friend of mine crazy is that so few people who "dance" in films can actually dance worth a damn. You can tell when there are constant cut aways and close ups of the feet (so they can sub in someone who actually can dance). When you look at the classic films with Fred Astaire, Gene Kelly, etc. they filmed with few edits and you see the whole body because they COULD dance.
 
Talking about dance on the screen. When you start to edit and not show the whole dancer uncut the magic is lost. Look at Fred Astaire or Gene Kelly. There are so few cuts and extremely well placed and doesn't interfere with the magic on screen.
 
Noiz2 said:
Of course on a sound related note the sound of his dancing was done in post on a Foley stage with his choreographer doing the tap sounds.
Click to expand...

You mean the sound from his "firecracker dance" (from the movie Holiday Inn) wasn't recorded while he was dancing? Damn, way to shatter my illusions. :grin:

Did you know the scene of Astaire "drunk dancing" (again, Holiday Inn) featured Astaire actually drunk? Rumor has it he took two shots of bourbon before they started, and one more before each successive take. They used take seven in the film.

He made it all look so effortless, but after nine shots it would take a lot of effort for me just to stay upright in my chair. Dancing? No frellin' way.
 
I had heard that somewhere I thought was reliable. A couple of shots would probably make me dance better but by nine I don't think I would still be standing let alone dancing.
 
Liam Hall said:
It's a four step process; EQ, Normalise, Compress, Normalise again. Simple.
Click to expand...

It is a FIVE step process.
If you don't start out CAPTURING it properly, no amount of post-production trickery will render really top-notch professional results.
It is a SERIOUS MISTAKE to think that this kind of quality is achieved in post-production.

Really well-recorded dialog will require minimal (or zero) post-production processing.
OTOH, poorly-recorded dialog cannot be saved with any amount of attempted magic.
You can't make a silk purse out of a sow's ear.
 
Richard Crowley said:
It is a FIVE step process.
If you don't start out CAPTURING it properly, no amount of post-production trickery will render really top-notch professional results.
It is a SERIOUS MISTAKE to think that this kind of quality is achieved in post-production.

Really well-recorded dialog will require minimal (or zero) post-production processing.
OTOH, poorly-recorded dialog cannot be saved with any amount of attempted magic.
You can't make a silk purse out of a sow's ear.
Click to expand...

Absolutely.
While the post steps will push it over the finish line - 80% of it is what you recorded. Totally agree.
 
Richard Crowley said:
It is a FIVE step process.
If you don't start out CAPTURING it properly, no amount of post-production trickery will render really top-notch professional results.
It is a SERIOUS MISTAKE to think that this kind of quality is achieved in post-production.

Really well-recorded dialog will require minimal (or zero) post-production processing.
OTOH, poorly-recorded dialog cannot be saved with any amount of attempted magic.
You can't make a silk purse out of a sow's ear.
Click to expand...

The OP was asking about POST...
 
Liam Hall said:
It's a four step process; EQ, Normalise, Compress, Normalise again. Simple.
Click to expand...

In decades of sound post I have never normalized a single vocal track, ever. And I can't think of a single dialog or ADR editor that has either.

Mic and actor choice would be #1. If it's post only then it completely depends on the actor and their voice. I find this "four step process" wrong on many counts. If you have recorded decent tracks there is no need to do anything like normalizing. If it needs some bump in level then do that with level automation. The only reason to ever go to a destructive level adjustment is if the tracks are FAR away from where they need to be. EQ is usually part of the chain. I rarely "compress" but I do use a limiter which is close to the same thing so either really will do. I mostly use compression/expansion in bands as a noise/ambience reduction tool. To some extent it depends on what media your working in. Radio ad's are often compressed to an extreme extent (as is a lot of pop music). Internet stuff is also often REALLY hot and compressed. Film work tends to be the opposite.
 
I generally do not normalize either. On occasion, I have 'peak' normalized to uniformally raise the overall level of improperly recorded file. I would never ever, ever, ever, ever normalize to RMS, which really has awful sounding leveling parameters.
 
Liam Hall said:
Don't knock it until you've tried it...
Click to expand...

I have "tried it". At best it is a blunt tool. It doesn't mix or even do much of anything positive for your sound. It doesn't do the damage today that early versions did to sound quality but it basically serves no purpose for dialog.

It was invented to maxemise resolution for samples in early low bit samplers. For that it works well. Some folks use it because they are just too lazy to mix. But it doesn't do that either it just raises the level everywhere. You can do the same thing with volume automation and not destructively alter your original sound. Limiting is useful and can basically combine a nondestructive normalize with compression.
You chain though goes through four destructive processes on your original audio, locking in things that you may well want to change later.

As a straight VO chain it's probably not awful since you are not balancing with other dialog, but you are still baking in choices and so largely defeating one of the big pluses of working on a DAW.

I look at that and see the look on the face of the director when they say "it sounds a bit too squashed" and you say " well go get some coffee while I find new originals and run them through the chain again...".

That usually means that IF I get to finish this job I won't have to worry about working with that director again.

As long as you are the one making all final decisions then whatever rocks you boat. I work in a world where I am rarely the final word and revision is the rule not the exception. And when they want changes they are often looking over your shoulder waiting for it to be done.
 
You must log in or register to reply here.
Back
Top