Some people think recording vocals just means putting up a mic, setting levels, and hitting record. But really, recording is a process—and it starts with the tracking.
With Zak Claxton scheduling is always tight, so part of the process is capturing as much material as possible during what little time we do have together. This forces an efficient workflow, but also one that takes multiple options into account, because it’s just not possible to have him come in and do a quick overdub later if needed.
I adjust the tracking methodology to suit the singer’s preferences. Some people like to go “old school” and track, come in and listen, and then decide what they want to punch in on to improve; others prefer to go until they feel they made a mistake, then, using a bit of pre-roll, continue on. Some singers prefer to do several uninterrupted passes, then comp, which is my default method. I like the “performance” aspect of it, and the uninterrupted flow often works well for the singer. But really, whatever makes them comfortable is best—I don’t want vocalists distracted by anything, so they can focus in on the performance.
As to gear, we started with a Soundelux ELUX 251 running into a Neve 8801 channel strip. I used a bit of compression on the Neve, but only few dB on the loudest peaks, and no EQ. A Stedman pop filter placed about 3–4" from the mic kept plosives under control.
PLAYLISTS: VOCALS UNDER THE MICROSCOPE
Each “pass” is done on a single track, but I use Pro Tools’ playlists feature and put each take onto a separate playlist of that track (Figure 1). In Pro Tools 8, you can then “fan” those playlists out so that they’re all visible simultaneously: Highlight an area of the song, and use the solo button on each playlist for auditioning the take. Once you decide which section you like, just highlight it, hit the arrow button, and it’s placed automatically into the main track. I compile to a new, empty playlist called “Ld Vocal Comp;” it’s a fast and efficient way of comping.
Note how some playlists are “full takes,” while others contain only certain sections of the song. As we do the passes, I make mental notes about any potential problem areas. If needed, we go back in and concentrate on fixing just those lines or phrases.
PIECING TOGETHER THE COMPOSITE VOCAL
After tracking comes finalizing the comp. There are several criteria in choosing the “best” performance; phrasing and pitch are important, but I also want to hear how the vocal builds, scrutinize the vibe and emotion, and make sure it will work with the phrases or sections that precede and follow it—you don’t want any formant or timbral shifts to give away the edit. I also listen for noises, such as mouth smacks and pops; if they can’t be edited out, I may pick another take for that section, or use an alternate take if something is far enough out of tune that pitch correction might be perceptible.
Figure 2 shows a screenshot of the final comp. Note the several different colored sections to the waveforms—these indicate which pass (playlist) they came from.
After the comp is done, I do manual crossfades among the sections. I could just select the whole track, hit Ctrl-F, and apply fades automatically; but manual editing prevents problems like having a fade between the breaths from two different recordings—this creates an unnatural-sounding “double breath.” I also make certain I’m not accidentally cutting off the beginning or end of any words, and while I’m at it, edit out any sections where the performer isn’t singing. But I never edit out all of the breaths: A singer who doesn’t breathe is unnatural. If any breaths are too loud, I’ll use volume automation to soften them.
Then it’s time to make the vocals fit into the mix . . . but that’s another story!