Vocal Cords: Turning Dreary Vocals into Pop Star Performances, Part II

Even after you’ve beaten a less-than-stellar lead vocal into submission, you may need to apply a little more polish before you can call it a day. I’ve found that a few single-line harmony parts can increase energy, as well as distract listeners from the main vocal. Although harmonies can be added anywhere throughout the song, I especially like to add these additional notes in the transitions from the verse to the chorus. 

Harmony synthesizers, pitch plug-ins, auto doublers, and other such processors offer infinite possibilities for creating harmony parts from an existing lead vocal. My personal preference is to do all my pitch modifications in Celemony Melodyne. To begin, I simply export a two-track mix of all the background vocals, a mono track of the vocals I will use to create the pitch variations, and a stereo mix of the music tracks. Once I open Melodyne, the first thing I do is set the session’s tempo to match the tempo of the incoming audio files. As with most DAWs, the tempo can easily be adjusted on the transport bar (Shift-Command-T on a Mac, or Shift-Alt-T on a PC). Once the tempo is set, I can now import the audio files I exported from Pro Tools to begin the modifications.

In Melodyne, there are a few things I always check to ensure the accuracy of my edits. The first thing I do is set the Scale Snap on, so I can properly analyze and transpose the vocals without any surprises. With the Scale Snap selected, I stop by the Prefs menu, and make sure the box that says “Consider original scale notes on Scale Snap” is checked. This enables Melodyne to understand which notes were—or were not—a part of the original scale. This is important, because as we generate new harmonies with different scale degrees, it will not force the original scale to snap to the new scales.

Now, I open up the arrangement view (Shift-Command-A on a Mac, or Shift-Alt-A on a PC), select the audio, go to Edit and choose Paste Special Menu, and then select “Copy-and-Paste selection to Parallel Track.” This will create a new track with an exact replication of the original audio file. Of course, this is going to introduce phasing, but, lucky for me, Melodyne makes this problem easy to fix. All I have to do is select the duplicated region, go to Edit, scroll down to Edit Pitch, and select “Add Random Offset to Pitch Center.” This randomizes the pitch center of the notes to simulate natural human fluctuations in pitch. This is one of the best ways I have found to ensure two copied audio regions have unique qualities. With the pitch center randomized, I now click and drag each note to the appropriate interval to create my harmony parts. With the new harmonies in place, I like to make a few minor adjustments to timing and formants. I do this to add a little more character to the vocal—making it even more different than the original file. As soon as I am pleased with my results, all that remains is to export my files and import them back into Pro Tools.

Feel Injection Via Whisper Tracks

The next time you listen to a song by Michael Jackson or Mariah Carey, pay attention to the techniques they use to increase the sense of passion in their words. You may notice these singers are masters of what I call “the breath to tone ratio.” A good ratio for this technique would be 70 percent tone to 30 percent breath. You want to hear the notes clearly, but you also want the intimacy of hearing the presence of breath (or air) change the feel and texture of the performance. There are five keys to recording successful “whisper tracks”: 

  1. The performer must perform the exact phrasing and intonation of the main vocal part. 
  2. They must sing as if they are whispering into the ear of a men’s magazine model like Vida Guerra, and that means no tone at all. 
  3. The performer must over annunciate the lyrics for maximum clarity.
  4. Keep the headphone levels down! The vocalist will be whispering at such a low volume that you will need to boost the gain on your mic preamp quite a bit. The last thing you want is to increase headphone bleed to a ratio of 60 percent headphone bleed and 40 percent whisper. That would make this whole process painful and pointless!
  5. Mute the lead vocals throughout the entire phase of the whisper track recording process. Believe me, you don’t want to deal with the challenge of editing out a lead vocal that slipped through the headphone defense system.

The actual recording process should pretty much be a breeze. The most important things to look out for are the little pops and clicks that a whispering mouth can make. Once I have stacked two stereo pairs of whisper tracks, I’m done.

Now that I have four amazingly recorded whisper tracks, I can begin the editing process. To begin, I solo each track for maximum focus. In addition to the quest to eliminate room noise—or any other glitches that slipped by me during the recording process—I do a simple “s,” “t,” and breath check for each whisper track. Especially at the end of a phrase, nothing sounds worse than eight out-of-sync “esses” that sound like a flutter of open hi-hats, 16 “t” sounds that have a sixteenth-note triplet feel, or 32 breaths that sound like an audience reacting to a scene. The goal here is to make sure timing and clarity are flawless in every performance. If the problems are minor, I use Synchro Arts VocAlign Pro. In life or death sessions, I prefer to perform each edit and time alignment manually, one “s,” “t,” or breath at a time.

With everything recorded and edited to perfection, I quickly solidify the presence of the whisper tracks with a few signal processors. At the top of the signal chain is GRM Tools Bandpass. As air transmits strongest in the midrange frequency, I bring focus to the mids by completely removing a lot of the top and bottom. Filtering out low end helps me eliminate as much room noise as possible, and cutting highs increases the clarity of the whisper tracks. Next, I slap on a Waves R-Comp, because I don’t want the compressor to react to anything other than the specified frequencies of my filtered vocals. I tend to go with extreme compression settings, which are great for helping whisper vocals kick in and provide maximum support to the main vocal. The next step is to “double” de-ess my vocals using two Waves DeEsser plug-ins. I set the plug-ins to two different frequencies—6kHz and 10kHz—mostly to catch any sneaky “sss” sounds that may have slipped through the filters at the top of my effects chain. Occasionally, I like to add the SoundToys Tremolator—no need to talk tech, it just sounds cool! Once my tracks are filtered, compressed to death, de-essed, and Tremolated, I send my “W-FX” track to a stereo aux return, where I enhance or alter the stereo field to taste using Waves S1 Imager.

Bringing It All Together

Because I work in various platforms, I do a lot of importing and exporting. It is absolutely imperative I remain extremely organized with every move I make. Once all the editing, comping, creation of harmonies, and imported MIDI tracks are brought back into my master session, I spend a lot more time fine-tuning every vocal throughout the entire song. Sometimes, making a song perfect means I need to undo one or two moves, or it could mean that I need to undo everything. The key to being a professional is being willing to do whatever it takes for as long as it takes to make something right. If you give your clients this level of commitment, the rewards and reputation you earn will follow you through your entire career.