OT - Final mix for radio spot

[Kent F]

Hey-

I'm looking for some pointers on levels, etc for final mix on a radio spot. Right now, the mix is not compressed other than using the levelizer peak limiter to knock off a few tops. No normalizing either. I know stations compress and I don't want to squash all the dynamics out of it before they do. What are you doing to prepare your final mix for radio?

If you're interested, you can download, evaluate, and comment about the wav (5.5 mb) at the following link:

http://whomadewhom.com/spot.html (Right click "Radio Spot" and "save as")

Thanks for your help.

[bcorkery]

I'm with Shats. Sometimes I EQ a notch into the music so I can mix it tighter. Peak Normalize enough that they ride up to about 97% and send it to the stations to get rally squashed.

[Ian Alexander]

This was fun to hear, because with the nice VO delivery and the acoustic guitar track, it's refreshingly different from most of the car and consumer show spots I mix.:)

I do a little of Stashu and a little of Bill.

I EQ the track, not the VO, as long as the VO has some presence. With today's sparkly mics, brightening the VO can make the overall mix a bit harsh to my ears. Depending on your VO talent, the place to back off the music 3-4 dB is typically between 5 and 7k, although another engineer was telling me today about a female talent who's S sounds are up around 10k.:eek: With your acoustic music track, EQ may not be necessary.

I usually go for some comp on the VO, to compete a bit with the typical squashed production music track. I set the comp for about 6 dB of gain reduction, 0.0 attack, and 100 ms release.

Get a decent mix on large and small speakers, because most radio listening is done in cars or on clock radios. If you are grooving to the music, back it off a bit. If the talent sounds like he's in the room with you, bring the music up a bit. Somewhere in between, it will "gel". Then "knock off a few tops" with the levelizer and max to 97.

I've never had or used a multiband comp, but the radio stations all use 'em, so they must be cool.:D

HTH

[Dave Labrecque]

Ian,

I, too, see female siblance up in the 10K region from time to time. Sometimes I think I'm more sensitive to siblance than others. I got the Levelizer doing it's de-ess thing quite often, usually after some compression of course.

What's really tough are the real broad-band (no pun intended) siblances, that go from, like, 3K - 10K. Yowsa! :)

[Demodave]

I am using AKG C3000 mics with light compression and EQ going in to SAW. Then, once in SAW, instead of a compressor on the vocal track, I use the dB-Audioware Limiter and a tad bit of Saw's Channel EQ to get the individual voice perfect. I might also add a touch of the Reverberator with a medium vocal plate depending on the spot. Many times...I don't add any more compression or limiting...not even on the output track. Although with concert spots, I tend to throw the Levelizer on the output track to juice up the entire mix and get consistent balance between the music hooks and the VO.

If you find that the music is almost too loud for the VO, notch out the 1K on the music by about -3 or -4 db and the adjacent EQ sliders by -1 db. This is much more effective than trying to EQ a great sounding VO to stand out above a busy music bed.

I hope this helps a bit! :)

[CurtZHP]

Yeah, I wouldn't go crazy with the compression. The station's air-chain processor will do that for you. Especially if it's a pop/rock station. They squash the snot out of everything until there's only about a couple dB of dynamic range (loud and louder).

[Kent F]

Thanks for all the input. I sure appreciate this forum and the helpful people.

Kent

[studio-c]

This is definitely NOT OT!! This is some of the most useful discussion we can have here, how to use SS to make our work sound better. Thanks for starting the thread!

Here's some fun stuff:

1- Use an aural exciter type plug-in as an aux effect on any vocal tracks. It will give them a 3 dimensional feel, sparkle, fairy dust, whatever, so they stand out from the music and sfx. Be very subtle though. (DSP-FX Aural Activator is an okay effect, and not too expensive. I'm sure there are others out there)

2- Do a mild sidechain limiter (ducker) on music tracks, which is kicked by the vocal tracks. The music will then subtly dip under the vocals and be hot everywhere else. Maybe kicking it down 5db or less. Play with the attack and release times, keeping them pretty tight. The SS channels are great for this.

3- Watch for places where the announcer goes soft, typically the end of sentences, and boost these so everything's even. Go through the spot syllable by syllable, looking for problem words that get swallowed or mumbled, etc.

4- Use the Levelizer to limit the mix mildly, maybe 2:1, the threshold being just the sticking up bits above the 'shoulder' of the main bulk. Most of the material will stay untouched. Hopefully this description describes the visual well enough for you. Play with this to taste.

5- In the Levelizer, use the Peak limiter to flat top the top 5 to 10 peak bits.

6- In the Levelizer, use the Normalize to take the level to 98 percent.

Save this whole thing as a template. Later you can drag your announcer lines, music beds, nat sound beds, incidental fx, etc onto the appropriate tracks and be halfway home with your next mix. Your ears are the final judge.

If you get the mastering done right, you can dumb-proof it enough to keep the radio stations from screwing it up.

Hope that helps.

Edit: I just read the siblance discussions. Learn to use the channel compressor as a de-esser. It will take you a half hour to get it down to a fine art, and will save you buying a plug-in. The Waves de-esser is pretty sweet though, I gotta admit. But try SS first and save bunch of dough :)

Also, on the levelizer topic, watch Bob's video. That's where I got all this stuff. Thanks Bob!!!

Also, if you can score a set of Auratone speakers, and get the mix decent on there, you're golden. Those things are hideous, and it's a great proving ground. :)

Cheers,
Scott

[DBenkert]

This is definitely NOT OT!! This is some of the most useful discussion we can have here, how to use SS to make our work sound better. Thanks for starting the thread!

2- Do a mild sidechain limiter (ducker) on music tracks, which is kicked by the vocal tracks. The music will then subtly dip under the vocals and be hot everywhere else. Maybe kicking it down 5db or less. Play with the attack and release times, keeping them pretty tight. The SS channels are great for this.
...
Edit: I just read the siblance discussions. Learn to use the channel compressor as a de-esser. It will take you a half hour to get it down to a fine art, and will save you buying a plug-in. The Waves de-esser is pretty sweet though, I gotta admit. But try SS first and save bunch of dough :)

Cheers,
Scott

Hey Scott (or anyone else),

Can you lay out a step-by-step how to on these 2 things? I know there are settings you can't tell because of different material we are using, but the general flow would be great. I currently use automation with slopes for "ducking" music and it works well, but I am trying to learn new tricks. My stuff is pretty simple. Generally a music track and 2 vocal tracks.

I think I remember a de-esser conversation on here before - maybe I should try to find that. If you can layout the general steps though, that would be sweet.

And I agree with all the Levelizer techniques you laid out - they work great!

Thanks!

David

[studio-c]

Can you lay out a step-by-step how to on these 2 things? I know there are settings you can't tell because of different material we are using, but the general flow would be great.
I think I remember a de-esser conversation on here before - maybe I should try to find that. If you can layout the general steps though, that would be sweet.

I'm at a location gig, so I'll do this from memory. Hope it's right :)

For the ducker, let's say your music is on track 1 and your vo is on track 2.
(It helps to label them in the title area on the left of the MT).
On the music track, in the compressor/gate area:
-Enable the compressor
-At the top, just below the eq, change the Key from "Self" to "Channel 2"
-At the compressor, start with Attack .010, Release at .100, and ratio at 1.5
These are not exact, I'm just thinking you want a pretty fast duck at the beginning of the sentence, and a mild release (tenth of a second) at the end.

Play your Multitrack with just these two tracks, and start edging the Threshold down till you start seeing the red indicators lighting up.

Then it's just finessing it to taste. Just play around till it feels right and doesn't sound like you can really hear it. It just gives you that tiny extra bit of clarity on the voice because it's not fighting.

Play with all of those settings and listen to what happens, then the numbers will mean something to you.

There are other discussions, particularly from BobL, about the de-esser. Again, just play around. Once you find that nasty spot, just Loop Playback and listen to the Key as you adjust the hi and low eq settings (like 5000 low and 5100 high) till you get that "AACK! That's the nasty sound!" You'll know it when it's right. Believe me :)



And I agree with all the Levelizer techniques you laid out - they work great!


It's a great tool. And all of those techniques are like figuring how much salt to drop into a pot of soup. Just mess with it till it tastes good.

Hope that helps!

Cheers,
Scott