The first time I heard Brian Free & Assurance’s 2007 album, Real Faith, my initial reaction was, “Man, that sounds almost TOO perfect!” The album, produced by Barry Weeks, was slick and highly-polished, but what really stood out to me were the vocals. They were obviously auto-tuned, but in addition, they were mixed (or, should I say, edited) to perfection; every single part was painstakingly adjusted to line up exactly with the other voice to the point that phrasing and even enunciation sounded 100% in sync, even if it wasn’t 100% natural.
Real Faith was the first album I’d heard using this technique, but it would not be the last, as “phrasing” has become just as common as “tuning” in studio language. Most albums now, thanks to a combination of enhanced digital recording techniques and ease of editing, consist of virtually flawless vocal performances, with an emphasis on the virtual part. It’s one thing to ensure that notes don’t hang over or cut off too short among four voices, but it’s another to replace natural performances with artificial perfection.
Phrasing, in and of itself, is not necessarily anything new in recording. For example, in the 1970’s, Donnie Sumner would have the support vocals (whoever wasn’t singing lead) sing only the vowel sounds of the lyrics to achieve the same basic effect – that of a single consonant sound that avoids any timing issues. Even that, however is done using natural abilities (even if it still creates a false sense of perfection).
Now don’t start flipping out on me saying, “What, do you want subpar vocal performances?” No, that’s not what I’m saying. I want NATURALLY good performances, not mediocre takes that are digitally enhanced. As I have stated in many past articles, I’d rather have a solid natural vocal with minor flaws than a questionable performance that’s been adjusted after the fact (sometimes to the point that it doesn’t even resemble the original take).
In fact, Joe Albano, an engineer and producer in New York City, opined in a recent article on Ask Audio that the natural imperfections are what creates a fuller sound. While discussing the benefits of doubling parts in the studio, Albano states, “…no matter how consistent a musician or singer is, there’ll always be enough subtle time and pitch variations to create the desired effect when the parts are layered.” He goes on to describe various methods used to create that doubling sound; among them are taking the same vocal performance and purposely making it IMPERFECT by adjusting the timing by milliseconds or modifying the pitch by a minor percentage. It’s the dissonance in the performance that creates multiple wavelengths, which takes up more physical space and gives the listener a larger (literally) sound.
But if we go back to the digital enhancements of modern recording, we’re actually going in the opposite direction. We’re still stacking SG vocals, but we’re tuning and phrasing them to the point that they sound virtually identical. Not only does this create virtually identical soundwaves, but those soundwaves actually step over each other, creating more of a phasing (“wah wah”) effect than a natural doubling sound. This phasing (not to be confused with phrasing) only adds to the artificial sound that’s created by an auto-tuner, and what we wind up with are overly-robotic vocals that not only sound fake, but are pretty much impossible to duplicate in a live setting.
And speaking of live settings….
Imagine you’ve just sung your vocal part on a song in a studio, and the producer or engineer says, “Ok, we got it, let’s move on.” You know the part you sang, you’ve sung it several times already, so you’re prepared to take it to the stage….that is, until you hear the final product; instead of hearing the part you actually sang, you hear a DIFFERENT part that’s been modified to fit perfectly with what the producer wants. You don’t remember singing it that way, because you DIDN’T sing it that way. It was edited later during mixing. Now you have to learn your part all over again, and it’s harder now than it was previously.
This is where producers both put too much work on their own shoulders and get lazy at the same time. Rather than work with a singer to ensure they have the correct part (and have sung it correctly), they hurriedly get one or two halfway-decent takes from a singer and move on, with the mindset of, “Ok, it’s close enough, I can fix it later.” The singer leaves the studio incorrectly thinking the have their part down (see above), and the producer (and/or engineer) now has to digitally alter the part once everyone else is gone. If you’re going to spend that much time on a part anyway, why not spend it with the vocalist instead of ProTools?
The answer, as all to often is the case, lies in the fact that most casual music fans either won’t notice these flaws (at least consciously) or don’t care. When you’re putting out a product to the market place, the goal is mass appeal, not catering to a select few, so as long as the mass market doesn’t care about manufactured perfection, then the artists, producers, and engineers won’t either. It’s more efficient and ultimately cheaper.
Part of the reason lies in the fact that studio recording has evolved so significantly over the last few decades that the old studio tricks are no longer necessary. Phil Spector used to hire two of every musician to record his “wall of sound;” now you just copy and paste the part. Sir George Martin achieved all kinds of sounds simply by adjusting microphone placement within Abbey Road Studios (using only a four-track machine and mix-downs) while James Taylor has an actual shipping container outside his studio for nothing more than creating reverb; all of those sounds can be duplicated using digital plugins. Like online ordering and voice-command technology in the home, modern recording has made life seemingly easier.
But like their other modern equivalents, easier recording techniques don’t always lead to natural results. You lose the personal connection with something that is artificial, and yes, manufactured.