Eq Dialogue Bus for TV drama

[nd33]

Hi guys
I'm doing the entire audio job on a comedy/drama show. I've done post audio on a couple of hundred shows over the past few years but this one is a step up in quality and alotted time for doing the audio. Hence I'm trying to do a really good job as opposed to making something passable within a minuscule timeframe.

I'm using 80% boom, 20% lapel. Generally it's ok recorded dialogue.

My queries are regarding getting a really clean sounding dialogue track.
I've been experimenting for the last few hours trying to get close to the tone of reference drama shows I regard as having pretty good production.

I'm finding with all my experimentation that I have made the most progress by using alot of LF cut. I'm talking 6-8 db low shelf from 500hz down.

Does this sound common for dialogue busses in general?

I've searched and searched for opinions on eqing dialogue but haven't come up with much relevant discussion.

There was a thread about "American sounding dialogue" and I tend to agree with there being a very smooth throaty ish kind of tight clean overall sound to the big money productions. It's not just American accents or speaking tone either as I live far far from there and I've heard it in a couple of local productions too, as well as nicely done shows from other countries. It doesn't matter which character is speaking or what gender they are, there's still a tone I haven't yet put my finger on. It seems to be a certain way of eqing mids as far as I can ascertain!

At the moment my dial bus chain is looking something like this.
Desser
Hpf 100hz
Low shelf -7db 500hz
Compressor 0-5db, 4ms att, 20ms dec, 10db knee, 2/1 ratio
Brick Limiter 0-4db
Lf boost 90hz +6db to add a bit of chest back
Lpf 10khz to smooth
Hi boost 10khz, med Q, to keep some clarity

As I said before, I really started getting closer to the reference when I started doing alot of cutting with the Lf shelf although I don't think I'm quite there yet.

Very interested to hear your comments/suggestions.

Thanks

[ggegan]

Originally Posted by nd33

At the moment my dial bus chain is looking something like this.
Desser
Hpf 100hz
Low shelf -7db 500hz
Compressor 0-5db, 4ms att, 20ms dec, 10db knee, 2/1 ratio
Brick Limiter 0-4db
Lf boost 90hz +6db to add a bit of chest back
Lpf 10khz to smooth
Hi boost 10khz, med Q, to keep some clarity

I would try to do more with less processing and keep things much simpler. I find it odd that you first cut frequencies and then later in the chain boost the same frequencies. Also, if you are going to have a limiter in the chain it should probably come last.

[BillBoy]

EQ's that you are doing a cut should come before compression and EQ's that are boosting come after compression, then brickwall limiter should come last. And I agree, why boost where you have cut? Keep it clean and compress it quite hard, with a proper compressor, this is what will make it sound 'pro' when mixed in with the rest.

[Henchman]

I don't go that low on my shelving.
And I don't use a master Dialog bus EQ to get my sound, I go clip by clip. Sometmes word by word.

[jahtao]

Originally Posted by BillBoy

EQ's that you are doing a cut should come before compression and EQ's that are boosting come after compression

No. EQ pre compressor. And compress post fader.

And yeah prob don't cut bass and then add bass?!?!? Just use simple, flexible EQ, so it sounds good

[nd33]

Originally Posted by ggegan

I would try to do more with less processing and keep things much simpler. I find it odd that you first cut frequencies and then later in the chain boost the same frequencies. Also, if you are going to have a limiter in the chain it should probably come last.

You're right it is a little odd! I guess that's because I'm feeling my way around to the tone I'm going after. When I'd done the low shelf cut it sounded alot closer to the reference, but maybe just a touch too thin so j added back some 100hz. What I should prob do is scrap the eq settings and start again

[nd33]

Originally Posted by BillBoy

EQ's that you are doing a cut should come before compression and EQ's that are boosting come after compression, then brickwall limiter should come last. And I agree, why boost where you have cut? Keep it clean and compress it quite hard, with a proper compressor, this is what will make it sound 'pro' when mixed in with the rest.

What do you mean "proper compressor"? I've just been using the digi rack3 plugin. I wonder if compressor choice has anything to do with the "throaty" sound I hear in the reference tv shows?

[nd33]

Originally Posted by Henchman

I don't go that low on my shelving.
And I don't use a master Dialog bus EQ to get my sound, I go clip by clip. Sometmes word by word.

I find that my large cuts in low mids are sounding pretty good on 90% of the show, with the remainder being a little too thin. Do you find yourself doing decent sized low mid cuts in MOST of the clips?

When you're eqing so many different clips or lines, how do you go about that? Audiosuiting or automating? If you're automating there must be a hell of alot of active parameters you're keeping track of in the session!

[Henchman]

Originally Posted by nd33

I find that my large cuts in low mids are sounding pretty good on 90% of the show, with the remainder being a little too thin. Do you find yourself doing decent sized low mid cuts in MOST of the clips?

When you're eqing so many different clips or lines, how do you go about that? Audiosuiting or automating? If you're automating there must be a hell of alot of active parameters you're keeping track of in the session!

I automate ALL parameters, all changes.
I try not to ever resort to audio sweetening, if I can help it.
And what is cut or boosted is entirely dependent on the sound of the clip.

[philper]

Originally Posted by Henchman

I automate ALL parameters, all changes.
I try not to ever resort to audio sweetening, if I can help it.
And what is cut or boosted is entirely dependent on the sound of the clip.

+1. Try to avoid overall EQ etc of any kind--production sound is too variable.
Only DeEss what you absolutely have to!! Start w/ the automation on a clip completely off, and see how little you can do to it to get it to work!

phil p

[ggegan]

Originally Posted by philper

Start w/ the automation on a clip completely off, and see how little you can do to it to get it to work!

phil p

My sentiments exactly!

[eoats]

I agree with Phil & the others, to a degree.

On each track I have a EQ3 7band & most often a De-Esser.

In most docs I make a dialog bus, which I will 1st put on a low cut on at 68Hz, 24db/octave & a Hi shelf at 8.5kHz, pulling it down 6db. (Digirack EQ3 7 band.)

Then a bit of De-Ess. Since everything seems better recorded & bright these days(sic); the Digirack set to the 'Female Dess Hi'.

After the EQ & De-Esser I use either Maxim or MasseyL2007 (often I 'finish' at various places and can't dictate every plug-in). Usually set to just peak limit at -17 dbFS, but with various delivery specs this changes.

[BillBoy]

[QUOTE=jahtao;5726386]No. EQ pre compressor. And compress post fader.

I disagree! Why would you boost frequencies to then have them compressed. You should see how it sounds out of the compressor before boosting because you could set the EQ before adding the compression and it will sound as you want it, then adding the compression can change this. The same reasoning goes behind why you cut pre compressor because there is no need in it going through the compressor to then be cut. But then, we all have our own ways of working so each to their own!

[rhumphries]

Just to add my salt to the soup... I start with HPF to cut out the rumble and mud, then some compression if necessary and then eq as necessary and I have a brick wall limiter on the bus. Sometimes a de-esser will be on the track too. I agree that I often end up cutting out a lot of low/low mids, but it all depends on the recording, so I cant tell you where and how much.

I also agree that the simpler the better. Ealier this week I spent quite some time trying to dig some dialogue out of a noisy recording only to discover that when I compared my work to the unaltered tracks, the originals sounded noisier, but better. I started from scratch & went back to some minimal processing and the tracks were better for it in the end.

-Richard

[philper]

Originally Posted by rhumphries

Just to add my salt to the soup... I start with HPF to cut out the rumble and mud, then some compression if necessary and then eq as necessary and I have a brick wall limiter on the bus. Sometimes a de-esser will be on the track too. I agree that I often end up cutting out a lot of low/low mids, but it all depends on the recording, so I cant tell you where and how much.

I also agree that the simpler the better. Ealier this week I spent quite some time trying to dig some dialogue out of a noisy recording only to discover that when I compared my work to the unaltered tracks, the originals sounded noisier, but better. I started from scratch & went back to some minimal processing and the tracks were better for it in the end.

-Richard

That's the classic audio-post merry-go-round! "Improve" the clip into a ruin, then go back to the original and discover you like it better!

phil p

[rhumphries]

Originally Posted by philper

That's the classic audio-post merry-go-round! "Improve" the clip into a ruin, then go back to the original and discover you like it better!

phil p

I'm glad it's not just me. I've had a bad few rides on that merry-go-round recently. It's time to step away and regroup.

-R

[danijel]

Originally Posted by nd33

Compressor 0-5db, 4ms att, 20ms dec, 10db knee, 2/1 ratio

This looks like a hell of a fast compressor - my starting point is usually 20-30ms for the attack and, like, a 150ms for release. I think you are killing the transients and making it sound flat that way. I really wonder what other people use as a staring point for comp. It also depends on the make, of course, but those are my starting points for both DynIII and Nuendo's fabulous, cleverly named 'compressor' plugin.

[Branko]

[QUOTE=BillBoy;5727935]

Originally Posted by jahtao

Why would you boost frequencies to then have them compressed

Because when you boost them, they may get out of control. Compression can be used to reduce the new dynamics of the program, as a result of messing with it's amplitude (EQ)
Just to add: general EQ has never worked for me. The beauty of ProTools is in total automation of every parameter. Use it.

Advice to original poster: instead of using a shelf and a boost for low mid, try using a wide parametric around 500Hz. You can get a better result, and it is simpler.

[jahtao]

Originally Posted by BillBoy

Why would you boost frequencies to then have them compressed? You should see how it sounds out of the compressor before boosting because you could set the EQ before adding the compression and it will sound as you want it, then adding the compression can change this. The same reasoning goes behind why you cut pre compressor because there is no need in it going through the compressor to then be cut.

This myth is a personal pet peeve of mine, sorry!!!!!!!!!!!!

I don't see why the compressor shouldn't hear the same frequencies you hear! (dialogue)

Is there any seasoned pro who ascribes to this rule of cutting pre-comp and only boosting post comp???!?

So sorry to get all fired up over such a small detail BillBoy : )

[Henchman]

Originally Posted by jahtao

Is there any seasoned pro who ascribes to this rule of cutting pre-comp and only boosting post comp???!?

Nope.
In fact I don't compress individual tracks at all.
In my Dialog master Chain, the last 2 things are a compressor and a limiter.
That's the only place I have compressors on my dialog..

[nd33]

Originally Posted by danijel

This looks like a hell of a fast compressor - my starting point is usually 20-30ms for the attack and, like, a 150ms for release. I think you are killing the transients and making it sound flat that way. I really wonder what other people use as a staring point for comp. It also depends on the make, of course, but those are my starting points for both DynIII and Nuendo's fabulous, cleverly named 'compressor' plugin.

I've done some further experimentation with adjusting attack time and knee and it has definitely taken me a step closer to getting the sound I have in my head. I think maybe I've underestimated the role of compression in getting the tone.

I'm using digirack3 compressor on the dial bus and have slowed my settings to 9ms (kinda peaky speaking accents so didn't want to slow it too much), 104ms release, 20db knee and 2.2/1 ratio. Compressing 1-6db generally.

What are you guys compressor settings? Any opinions on the role of the compressor in your dialogue tone?


Thanks so much for your input fullas, I've definitely learned a few things

[jahtao]

Originally Posted by nd33

I'm using digirack3 compressor on the dial bus and have slowed my settings to 9ms (kinda peaky speaking accents so didn't want to slow it too much), 104ms release, 20db knee and 2.2/1 ratio. Compressing 1-6db generally.

Cool! That's pretty much exactly what I use! Tho' for dramas I have been known to mix without compression or to ease the ratio to 1.5:1

Comp goes on the dial bus!!!

[jahtao]

Originally Posted by Henchman

Nope.
In fact I don't compress individual tracks at all.
In my Dialog master Chain, the last 2 things are a compressor and a limiter.
That's the only place I have compressors on my dialog..

So you do compress! I think I had got the impression (from older threads) that you were anti-compression all together.

Nice.

[longblack]

Originally Posted by nd33

Hi guys
It's not just American accents or speaking tone either as I live far far from there and I've heard it in a couple of local productions too, as well as nicely done shows from other countries.

As another kiwi , just curious which local shows you have heard this pleasing tonality on?

[Henchman]

Originally Posted by jahtao

So you do compress! I think I had got the impression (from older threads) that you were anti-compression all together.

Nice.

Yes, I do. But have the threshold set very hi, so it just rounds it off before it hits the limiter.

Ie. For the limiter, if my spec is -3 db, then I'll set the dialog stem limiter to hit at - 3.5

[nd33]

Originally Posted by longblack

As another kiwi , just curious which local shows you have heard this pleasing tonality on?

I've just been watching Outrageous Fortune on DVD.
I know from talking to a couple of peeps that it is a rush job (aren't they all?) but it still has that familiar throaty tone that cuts through so well on tv.

I think I'm definitely closer after cutting more in the 200-500hz range and slowing down my attack and release settings on compressor.