Production 212: Mixing In A New York Minute

Production-212-Logo-1By Dave Foxx

Y’know, sometimes an idea has to conk me on the noggin a few times before I realize what it is. Case in point: On three different occasions over the last few weeks, as I traveled to show people how to do this voodoo, I’ve been a bit stunned to see people not conditioning their audio before they start producing. They just start throwing stuff up on the edit window and moving it around, and then spend a lot of wasted time (in my opinion) trying to make the mix work. After I carefully explained the concept of conditioning the first time in Toronto, I went on my merry way, not thinking about it again… until I saw the same thing pop up in another session in Los Angeles. Finally, after a lengthy discussion about it (my third in three weeks) with one of our web audio producers here in New York, it occurred to me that not everybody has figured this out yet. In the off chance some of you haven’t figured it out, I decided that this column was the perfect forum for this bit of sage advice.

Let’s begin with a bold statement: Not all audio is created equal. Think about it for a moment. In one promo, I might use two or even three voice tracks that were all recorded in different studios, Los Angeles, Chicago and New York. I’ll probably use two or three pieces of hit music, which again might have all been recorded in different studios and mastered by different producers. Add the various and sundry effects and beat tracks being recorded all over the world, and you have a lot of sounds being dumped into one session that is supposed to make them all play harmoniously, seamlessly in fact, to help me get across my message.

Well, you can simply throw them up there and cut and paste, nudge and stretch to get them all in a semblance of order. A good producer can then go to work on the levels, adding a bit of compression here, and some EQ there to get it all to come out right. BUT, my friends, it’s a complex and sometimes tedious chore to make the mix work. I know. I’ve done it many, many times. Several years ago, I had an epiphany. If you simply go through all the various pieces of audio with what I call a conditioning process before you begin the edit, you can save yourself a ton of work. Once you get used to doing it on a regular (read every time) basis, you will discover that the mixing process takes mere moments. You will breeze through the back end of your project without even breaking a sweat.

Here’s how it works.

Begin with the voice tracks. If I am using all 3 of the station ‘voices,’ Kelly Kelly Kelly, Dave Kampel and myself, I throw them all up (unedited) and go through some basic processing to make them all sound like they were recorded in the same studio, one right after the other. If one has more compression than the others (usually mine), I use that as the baseline, adding compression to Kelly and Dave’s tracks to make them punch as strongly as mine does. I create new files with the compression built in, which I will then edit and use in the session. (On a side note, having a visual component to this makes it even easier. Since I use a clipping limiter to effect the compression, once the waveforms look similar, I know they’re beginning to sound similar.) I will also add some high pass EQ to their tracks to, again, make them sound like mine. Kelly and Dave both send filtered versions of their settings, so the need for this is minimal in this case. Bear in mind, this entire process takes a couple of minutes, at most. This alone has the potential to save dozens, if not scores, of minutes in the mixing process.

Most of the time, I do NOT use more than my own voice but, having done this more than a few times, I’ve learned to go through this process anyway, just to make sure I sound the same all the way through the read. It still saves me time in the end.

Step two involves the hit music I’ll be using. I try to find instrumental versions of the songs. Usually, I can find them online at Reel World’s Production Vault. (Some songs simply do not have instrumental versions available unless you have strong pull at the label. Of course, you need enough lead-time to get one of those tracks.) The instrumentals generally are a bit stronger than the full vocal versions, mainly because when the vocals are gone, producers turn up the gain a bit to compensate. So I will normalize them back down to the levels present in the full mix. I match up the instrumental to the full mix so that they’re ‘mostly’ in sync, and then hunt down the hook, slice and dice so that as the final version plays, the only vocal IS the hook. I never ever allow talk over the vocal. The only other option is to cut out of the vocal into a completely new piece of music. Again, this entire process takes a couple of minutes, but it makes doing the final mix a piece of cake.

Finally, I go through all of the music tracks, hits or not, and get them all to sound like the same producer/engineer mastered each of them in the same studio. This requires a bit more finesse than the voice tracks. I am loathe to compress hit music beyond what the original has, so I tend to use more expansion, combined with EQ to match them up as much as I can, although I never take compression fully off the table. (Try working with a Jason Mraz song and a Rihanna song in the same promo without some compression. HAH!) Included in this step are all the specialized production tracks you’ll be using. To help a bit going in, I tend to NOT mix libraries. If I’m going to use one of the Trynity HD/FX cinema tracks, I doubt I can even come close to matching that with something from Chase Cuts or Production Vault without some major fudge-factor. So, if I start with Trynity, I stick with Trynity.

Now… everything has been conditioned. I edit the VO, start sliding the music in and around the reads and adjusting the timing of everything so that it feels right. If I need more time from the VO track, I’ll consider stretching it, or perhaps adding a quick drop like “Let’s GO!” to make everything time to the music tracks. At this point, I take a kind of ‘holistic’ approach to my work and adjust timing as I make the music ‘fit’ together on the rhythm and tempo. Instead of asking, “which comes first, the music or the voice,” I tend to put the music first, but it’s a delicate ballet that often works the other way.

NOW, it’s time for the mix. Oh, look at that. Because everything sounds like it came from the same recording session, all I need to do is a bit of ducking whenever the music overpowers the voice. I’m done. It’s ready for air and time for me to head for the kitchen for another cup of coffee as I ponder my next project. Instead of spending 40 to 50 minutes on the mix, I spent 5 to 10 minutes on the conditioning. Not only does it save a lot of time over the course of a day, my work sounds better because I basically did the mix before I did the edit.

If you’re inclined to try using this method, which I strongly suggest you do, give it some time to start working for you. I would guess that you will probably curse me a few times while you struggle to get your conditioning squared away, but you’ll smile when you do the mix. Like anything, it takes practice to become really proficient, so give it a month… then pat yourself on the back for learning how to put the cart before the horse. The view is really much better that way.

The audio for this month is a very simple, yet powerful promo we ran the third weekend in August. I don’t mean to brag (yes, I do) but it only took me 18 minutes to produce it, from laying down the voice track to heading for the kitchen for more coffee. The mix? Ever heard of a New York Minute?

Comments (0)

There are no comments posted here yet