Recording Vocals with Compression
My first experiences with vocal compression left me a mite ambivalent.
On one hand, the idea of a vocal already being recorded with very minimal volume fluctuations was so exciting that I’d lay awake all night anticipating the benefits I’d reap from purchasing my own vocal compressor. But, on the other hand, I would sometimes hear negative feedback about compressors from other recording enthusiasts.
In a moment, I’ll give my final view on vocal compression, along with specific settings that some may find useful to start with on their own compressors.
For now, though, I hear Memory Lane calling me. In 1997, I began frequenting a local recording studio, laying my vocals over Karaoke tracks. The owners of the studio took a liking to me, and offered me a job. It wasn’t the most glamorous of jobs, but I was ecstatic.
In one week, I’d gone from being a Kroger courtesy clerk, to working in a recording studio! There’s absolutely nothing wrong with being a courtesy clerk. But, on hot summer days, wouldn’t most anyone rather be in an air conditioned office packaging CDs, than to be out in the miserable heat with grumpy customers warning you not to mash their bread before you even make it to their car?
Packaging CDs wasn’t my only duty at the studio. I also took on the unofficial janitorial duties. But, as time went on and I proved myself as a hard worker with an interest in learning, I began being invited to the control room more and more.
One day, the engineer and I were listening to an isolated vocal track.
A customer had brought it into the studio to hopefully repair the manner in which it had been recorded at the vocalist’s home studio. The vocal belonged to a wonderful singer… but, the recording process of the vocal was repulsive, even to my then novice ears.
“So, Ben, why does it sound so… weird? It’s like I hear him taking big breaths between every phrase, and the breaths are louder than the singing.” Ben explained to me that the vocal had been recorded “too vigorously through a compressor”.
This puzzled me, because while I had still been just a client, Ben himself had recorded me with some compression on the way in. His reason had been that I had good dynamic control, but for the overall mix of the song, he just wanted to snip off a few sudden “peaks” in my voice before the millisecond that I pulled my own volume down naturally.
The recording studio began having some turbulent financial times, and sometimes employees didn’t get paid. I moved on, which was a hard decision. I needed a dependable income, and yet I was thirsty for the knowledge to be learned in that atmosphere. I found another job, saved up $150 and eventually purchased my first multitrack recorder. Those were some fun times!
On my days off from work, I’d spend up to 12 hours or more singing until I was hoarse.
I’d obsessively “ride the fader” of my vocal during playback, trying to keep the volume even throughout the song.
When I became a stay-at-home parent in 2003, the act of having to “ride the fader” of my vocal tracks became a frustration. Taking the time to sing was simple. My son loved my singing from his first weeks of life, and when it was time to give him a nap, I’d lay him down in his bassinet right beside my workstation. But, when he’d wake up, I would stop what I was doing and fight preoccupation over how well I’d sung my vocals.
Only by riding the faders and seeing how my voice “sat” in a song could I know how good a vocal performance really was. The internal catalog of my mind pulled up some data about past thoughts on compressors.
I must have stood in Guitar Center for an hour, debating whether or not to buy my first Alesis 3630 compressor. The store assistant told me that the Alesis 3630 was “a very noisy unit”. A fellow customer overhearing us told me that the Alesis 3630 “is a very good unit as long as someone knows what they’re doing.” I got home, and handled the compressor like an unsupervised child.
“The attack knob controls how fast my loud notes get turned down? Let’s go to the maximum!” “Ohhh, the release knob determines how long the loud note stays clamped down? Maximum!” “A 2:1 ratio only lowers my loud notes by 2 decibels? Let’s go for a 6:1 ratio!!”
I would sing my heart out, positive that I’d just play back my songs and it would be almost as if the songs had already been mixed for me. That false optimism made me sing better than I ever had! “Wally? Uhm… your voice is pretty, as always. But, you sound tired or lazy or something. I can’t put my finger on it.” My wife really deflated me that night.
I knew beyond all doubt that I was singing far above my own expectations, and she tells me that I sound tired and lazy? Discouraged, and with increasing clientele for my graphic design business, I put away the recording equipment for a while.
A less hectic 2008 found me yearning to record again. Reading eases my occasional insomnia, so one night I chose to read about the subject of vocal compression.
I learned a lot that night that I wished I’d taken time to research in 2003.
That attack time, which I’d been too eager to use? It needed to be set a bit higher so that just the first blast of my loud notes would still be heard before the volume got lowered.
I’d done even worse with the release knob. The lower it is, the more “artificial” a signal begins to sound. If the attack and release knobs were set so badly, what about the ratio? 2:1 is usually substantial to just take that loud edge off of a vocal peak.
I began recording again, and still just couldn’t find the perfect setting that would free me from having to “ride the fader”. One day, I recorded a plain, uncompressed vocal into my laptop sound card. Utilizing what is known as a VST plug-in, I decided to try compressing my vocal digitally. This particular tool was the Kjaerhus Classic Compressor.
I finally heard a closer sound to what I had been listening for, simply by using the “vocal” preset.
The on screen knobs were kind of vague to read, but I did manage to find the script for the presets.
Those presets, along with some very minimal tweaking, finally brought me to my long awaited “A ha!” moment.
Today, I still use the same Alesis 3630 compressor. Am I able to sit back and just listen now, without having to “ride the fader”? Sometimes. There may be an occasional softly sung note, or a really bassy note that I have to manually pull up. But, this is not very often.
The Guitar Center employee who told me that this was a noisy unit has been proven wrong.
In fact, I find that the signal to noise ratio is much cleaner with the compressor than when I plug my microphone preamp straight into a recorder. I’ve had other recordists relate terrible Alesis 3630 experiences to me. Some claim that no matter what settings they use, their tracks just sound “processed and unnatural”.
On an opposite end, I had a recording engineer once remark about one of my songs that it was “nice to hear someone record a vocal without any compression and use the vocal fader the old fashioned way”. I sheepishly admitted to him that I’d used the industry cheapo… Alesis 3630. He thought I was joking.
To compress, or not to compress?
My thoughts are that some very minimal compression is needed for most vocals, unless the singer has the world’s best dynamic control to ever exist and also never exceeds a certain volume.
Even an untrained ear can hear the blaring, loudly recorded notes on Celine Dion’s 1996 hit, “It’s All Coming Back to Me Now”. The blaring is most evident on “more” and “laws”, as she belts, “it was more than any laws allow”.
Celine Dion possesses one of the most beautiful voices to ever be heard in this universe.
And, as such, she was recorded by a top record label with top notch equipment.
So, some notes by any vocalist are just sometimes going to be loud. I have tried to research and find if Celine’s vocal on that song was recorded with or without compression. I have had no luck in uncovering this information. I would almost guess that it was recorded without compression.
I have heard notes of an uncompressed vocal take sound distorted even without any fathomable cause. Even with no “red light” to indicate microphone clipping.
The same delivery of the distorted note, on a different take, recorded through a compressor, would record cleanly, with absolutely no distortion.
It just all depends on the sound that a person is going for. I prefer what I call “natural, vintage vocals”. My vocals are big and up front just enough for clarity amidst the music.
The volume is basically the same throughout the song, but that occasional “Hey, pay attention!” loud note will stand out just a little bit more than the softer notes.
Obsessive Compulsive Disorder makes me check, every time I am about to record, that my Alesis 3630 knobs haven’t been moved since their last use. Peak mode, as opposed to RMS. RMS mode can sound more natural, but it doesn’t soften peaks the same way that… well… peak mode does.
Is my threshold set to -15 db? It should be, because this means that there’s always some small degree of compression going on, with -15 db being my average, medium volume level.
The ratio knob has hopefully been left at 2:1. This means that if a loud note exceeds -15 db, it will be pulled down, but only by a very natural 2 decibels lower.
I want the attack knob to allow the very front of my loud notes to remain intact.
The release time, I’m hoping, is still set nicely between fast and medium.
After all of this volume lowering, my output needs to stay increased to about +8 db.
Wait! Is everything set to the more musical “soft knee” mode, instead of “hard knee”?
The only drawback I find in the Alesis 3630 is that the in-betweens of the knob presets are blank, and there is no way to set things to an exact number between these presets.
As far as sound quality goes, I have tested many brands of compressors over the past few years. To my picky, critical ears, the $99 Alesis 3630 keeps the integrity of a performance as pristinely as its $500 competitors.
As long as any compressor that I’m using looks something like the drawing below, I know I’m going to have a very natural, non blaring vocal, with minimal need to “ride the fader”.
Walvis struggles with Obsessive Compulsive Disorder minus the aid of medication… but, don’t worry, he frantically Googles the same keywords every Monday at 8 A.M. in hopes of learning new coping mechanisms for said OCD.