Vocal Mixing Techniques: Automating EQ for the Ultimate Ride
With any technology comes a list of the pros and cons – the benefits and the detriments. (“Is that a flaw or a feature”)?
Certainly recording and mixing “In the box” has brought a full complement of both halves of that list, and today I want to talk about one of the greatest, and least talked about, advantages of DAW mixing – fully automatable EQ.
Simple EQ automation of instruments that share a track is common, but when I really want to ride a vocal, I do it with EQ and level.
After I get my basic sound – level, eq, compression, delay, reverb/echo, and panning, I will do a few small rides to make sure the vocal rides the crest just where I think it should against the track from top to bottom.
Then I begin the process of fine tuning the performance. This may include tuning if the client/singer doesn’t mind. A big part of the final vocal placement comes with the rides, and for that I set myself up with a great eq or channel strip and automate several if not all of the parameters.
My four favorite devices for this are: Brainworx bx_Hybrid 1.0 EQ, Massenburg MDW 5 EQ, the Waves SSL Channel Strip (or the UAD version of the SSL), and the McDSP FilterBank E6 . I like the Brainworx a lot for both the graphic display of the ever-moving curve, and for the auto-listen mode for zero-ing in on offensive (or pleasing) frequencies.
Even a well recorded and well performed vocal track can use EQ adjustments, frequently for taking care of pops and bumps, but also for subtle proximity effect balancing and especially for ear-piercing hi-end passages. You know, for the times when the vocalist sings nothing like they did for any of the run-throughs!
Never in the analog world were we able to make real-time sweeping EQ changes during a mix, and I find it to be one of the most satisfying elements of the digital recording age – right up there alongside non-destructive punch-ins!
As the songs plays I might adjust the low end down (in shelf mode) for the verses when the singer is snuggled up to the mic and singing in their lower register softly. The frequency point of the shelf is usually around 150-250 cycles. As the song intensity grows and builds to the chorus, often the singer’s timbre and mic technique change radically, and I have to put the bottom end back, and take away (in a very narrow Q setting kind of way) annoying high frequencies. Believe it or not, for females this is usually 9-10k NARROW. All the while riding the output gain knob of the EQ to make up for what I may be losing (or gaining).
Maybe the midrange is too intense for the Bridge, but fine everywhere else. You get the idea – I’m riding not just overall levels, but specific levels that taken together have the effect of a more subtle and even natural vocal presence.
Not just for human variations, but for technical ones too – in the age of the endless vocal comp, we sometimes deal with vocal performances that were made days, weeks, months, even years apart! With a fully automated channel strip I can handle any sound discrepancies with pinpoint accuracy. It also happens to be a ton of fun to do.
This is especially true for me, as I handle all these moves with faders and knobs on my ICON (DControl) console in Plug-In Map mode. I much prefer the tactile experience of moving faders and twisting knobs to the laborious effort that would be the alternative in the mouse and click world.
A classically trained musician, Brian became immersed in sound engineering in 1980. By the end of that decade he had amassed a string of hit albums, singles, and soundtracks for artists such as Michael Jackson, Madonna, Prince and Roxy Music. In the mid-90’s he joined RCA records as A&R/staff producer and helped shape the careers of Everclear, The Verve Pipe, Eve 6, and Lit, among others. In July 2004 he joined Columbia Records, and in ‘08 Disney. Today he is an independent mixer/producer, serves on the Board of NARAS and co-chairs that organization’s LA Chapter of the Producer’s and Engineers Wing.