The Star Treatment

Lead + Harmony Vocals

By Charles Dye

	Fritz, Eric, Greg and Kelly of CREASE

Click here to listen to Live To Be In Love with all the tracks, including the final lead and harmony vocal tracks. We still have more tweaks to do, and the automation so far is very basic. But this is how a mix would normally sound when I first present it to a client or artist for feedback.

Which Came First: The Vocal or The Track?
If the lead vocal is so important, why do we wait until the end of the mix to put it in? Well, not everybody does. Some mixers get the sound for their lead vocal first. They may spend an hour or more listening to only the vocal. They'll EQ, compress, and put effects on it before ever bringing up another instrument. The concept is to get the lead vocal sounding great and then fit all the other instruments around it. Mixers who prefer this approach feel that it is easier than getting sounds on all the other tracks and trying to then find a place to fit the vocal in afterwards. It apparently works best on music styles that are vocal-driven. If this applies to the music you are mixing, you may want to try this approach.

Kelly Meister singing lead vocals for Live To Be In Love

I use the track first approach (you may have guessed), but I understand the logic of the other technique. In my early mixes I would sometimes make the track sound so full or complete without the lead that I had a hard time trying to squeeze it in afterwards without making a lot of changes. But after a while I learned to project in my mind where the lead vocal would be and leave room for it in the other tracks. With panning, I now keep most of the instruments outside of +/-35 or 40 so it doesn't crowd the vocal. And with EQ, I allow room for the vocal's fundamental frequency, which is usually in the 100-250 Hz range. I don't use an overabundance of it in the rest of the tracks in order to leave some space in the high-mids for the vocal's presence. I also constantly check the rest of the instruments against the lead as I'm getting their sounds to make sure that I am leaving room for it.

Another reason I use the track-first/vocal-last approach is because of the way I treat the vocal in direct response to the track that will be riding beneath it. My vocal reverb, delays, compression and EQ decisions are all based upon what the other tracks are doing. I don't think I could make these choices within the vacuum of a soloed vocal. I need the track to guide me. But I'm sure that just as I project where the vocal will be when working on the kick, vocal-first mixers simply do the reverse.

New Delays for the Vocals
When mixing, I essentially have two sets of delays — one for the instruments and another for the vocals. The difference is usually in the treatment the delays have on their returns. My instrument delays may be drier or have smaller reverbs, and possibly less feedback to keep them from cluttering the notes of the other instruments. The vocal delays will usually have similar effects, but with more of them, and probably an additional larger reverb (maybe a Hall) to make the delay seem further in the back. We discussed DDLs 1, 2 and 3 in HDL 7 that we used for the instruments, below are our vocal delays.

DDL4 — 1/8 Vocal
Again, I use the DigiRack delays. The reason is that I like the vocal delays to appear more in the back and not to distract from the original vocal, while still adding to the space and size of the mix. I roll off even more top-end than I do with the instrument delays and I generally set the feedback on eighth note delays to around 7 to 13%. The other effects I use are the DPP-1 Split Harmonizer effect, and the LexiVerb 1 Vocal Plate setting, with more effect in the mix than the vocal track.

DDL5 — 1/4 Vocal
I set the quarter note delay similarly, but with more feedback (13% to 35%). I'll usually add some of the Reverb One with a Large Hall setting.

The Tracks
We recorded Kelly's vocals at The Gate in Miami and AVC Studios in Ft. Lauderdale. We re-recorded much of the original reference vocal, but we did keep some of it because it had the best feel. So the final vocal comp is a combination of the two.

Below, I'll go through each of the vocal tracks and discuss my reasons for using the plug-ins and settings I did. For the exact settings, see the screenshots or download the session from DigiStudio. To download just the session data without the audio files, set your Audio Receive Quality to None. (See HDL 8 for more details.)

The Lead Vocal
What actually defines when the vocal just sounds good and when the singer sounds like a star? Is it all up to what's already on disk? Definitely not. I learned this when I was doing a rough mix of a soon to be big fat hit and my knob tweaking on the lead vocal just wasn't cutting it.

So, what is a star anyway? A star is bigger than life. She is perfect in her imperfections and someone the audience can relate to, connect with, and feel for. Some of this must be taken care of before you begin the mix. A good performance that has captured the singer's emotions as she connects with the lyric is important. Now, all you need to do is connect those emotions with the listener, and keep the listener connected.

Ren Comp
The first plug-in we are using is one of the best in helping us connect the singer to the listener. By keeping the vocal's dynamics more even with the compressor we can keep all the nuances of the singer's performance in the forefront without it overpowering the track.

I use settings similar to the compressor on the BGs in HDL 9. I prefer the Opto and Warm settings usually on vocals. I will compress the vocals as much as is appropriate depending on the song's tempo and how much the vocal is compressed. I go all the way from 2 dB to 10 dB of attenuation. I find that faster-tempo songs often require more compression to keep the vocal upfront, and slower songs require less. And regardless of tempo, vocals need more compression with denser arrangements and less with tracks that are more open.

De-Esser
The de-esser is very important with lead vocals. Bringing out the vocal's air (as well as some of its emotion) generally requires a lot of top-end EQ, which makes the esses so bright they sometimes will shave off your sideburns. To combat this, a de-esser cuts down on the esses and other bright consonants while leaving the brightness of the vowels. And that's how I set it — in such a way that it only attenuates the consonants. I set it essentially the same as I did the de-esser on the BGs in HDL 9.

Ren EQ 6
The six bands of Waves' Renaissance EQ I use are all for different areas of the vocal; the HPF is clearly used to roll-off any unwanted low-end; the 100 Hz range is designed for the vocal's fundamental; the 280 Hz range is to bring out warmth; the 2.2 kHz area is for mid-range cut and intelligibility; the 5 kHz area is for presence; and the 16 kHz register is for vocal air.I generally prefer a narrow Q of around 1.3.

I EQ the vocal a little bright in anticipation of the next plug-in — DaD Tape — which rounds off some of the top, and it brings out some low-mids. Because of this, I sometimes come back to the EQ after inserting Tape and back off some of it as well.

	The plug-ins and settings used on the Lead Vocals. With the exception of not using DaD Tape on the Harmonies, the plugs and settings are all very similar for all the vocal tracks. (click to enlarge)

I have to admit I stumbled onto this completely by accident. I wouldn't have done it intentionally, but nonetheless I did and the results are really great. It adds a nice saturation to the vocal for sure, but that's not why I like it at all. The end result is essentially automatic vocal rides. We'll discuss vocal rides when we get to automation. Suffice it to say they are essential to giving the vocal that star quality, and also they require a lot of work to do properly. Tape seems to do a very good job of keeping every syllable of the vocal up front, giving the it the almost effortless sensation of always being close to the audience, and without all the work.

Aux Sends
Here are the effects that give us the larger-than-life qualities. It's important to give the singer an ambience that fits the mood of the track, but be careful with effects — too much can actually hide the emotions from the listener.

> DPP-1 — Split Harm. This is the first of the bigger-than-life plugs. I generally set it to around -19.0. I don't want to really be able to hear the effect, just to sense that it is missing when I mute it. It has a subtle affect. But the result is that instead of the singer sounding like he or she is coming out of the middle of the stereo field, the singer actually becomes the stereo field. With this effect, the singer is now coming from the left, right and center. This setting also adds a nice richness to vocals, and a bit of thickening without sounding electronic.

> LexiVerb 1 — Vocal Plate. This somewhat bright plate is great for giving the singer a space that brings out the top end. You don't need too much, just enough to make the artist not feel naked in the track.

> DDL5 — 1/4 Vocal. This is a real larger-than-life effect. Not an obvious delay that is constantly ringing in the background, just a small amount to give the illusion that the singer is in a large space and the voice is projecting all the way to the back wall and echoing off of it. It often sounds great in choruses.

> DDL4 — 1/8 Vocal. Sometimes I prefer the eighth note in verses of songs instead of the quarter note. It's appropriate, because verses are usually sung in a smaller voice than choruses. Again, don't use too much or you'll distract from the performance.

Listen to samples of the Lead Vocal
Before | After

Lead Double
This track adds a thickness to the lead when it comes in. I usually keep it panned in the center, though sometimes I may pan it slightly off (+/-15), and I usually keep it about -6 dB under the lead. All the plugs and sends are set essentially the same as the lead. A possible exception might be to make the EQ a little less bright and add just a bit more of all the effect sends. I want the double to sit a little behind the lead. This is not a hard and fast rule.

Ren Comp -> De-Esser -> Ren EQ 6 -> DaD Tape
See above.

Aux Sends
See above.
> DPP-1 — Split Harm.
> LexiVerb 1 — Vocal Plate.
> DDL5 — 1/4 Vocal.
> DDL4 — 1/8 Vocal.

Listen to samples of the Lead Double
Before | After

Harmony Vocals
Again, I treat the harmonies pretty much the same as I treat the lead double. But I pan them relatively wide (+/-50 to 100) and I EQ them in an effected way if appropriate.

Ren Comp -> De-Esser -> Ren EQ 4
See above.

Aux Sends
See above.
> DPP-1 — Split Harm.
> LexiVerb 1 — Vocal Plate.
> Reverb One 1 — Large Hall. I like to add a bit of the Hall to move the harmonies to the back.
> DDL4 — 1/8 Vocal.
> DDL5 — 1/4 Vocal.

Listen to samples of the Harmony Vocals
Before | After

Now that we have all our tracks, we will take a look at The Big Picture next month: balance, blend, F/X, and groove. We'll bring them all together so the song makes sense. We'll also examine stereo bus processing, otherwise known as the "record button" (as in long-playing record).

Engineer / Mixer / Producer Charles Dye (Jon Bon Jovi, Sammy Hagar, Ricky Martin, Ednita Nazario, Julio Iglesias, Robi Draco Rosa) is based out of Miami. www.charlesdye.com email: charles@harddisklife.com

©2003 Charles Dye