Okay, so here is the issue I'm having. When I'm recording an artist for hip hop, and they want me to mix a vocal harmony, I typically do three (3) to five (5) takes on the recording, to give it richness and depth (depending on the strength of the vocal take).
My questions relate to the above, and are thus:
1. How do you typically pan your vocal takes to increase the dynamic body of the harmony?
2. What is more important when layering, to you, delay or reverb?
3. Do you prefer coaching your artist (vocalist) to perform the key, or re-tuning the vocal takes?
GO!
If I have a vocal harmony that works with the stereo element to create a nice wall of harmonic vibe, then one thing I do is to leave some space between those sound sources and the other ones in terms of pan width, because then I can also pan the vocal harmony a bit less than 100%, which fattens up the outer edge of the stereo image by adding some frequencies on the opposite side and leaves some room for delay and reverb most far out.
Delay and reverb must be timed properly so that it does not cloud/mess with the frequencies in the mix. I prefer to add reverb behind the delay, so that when the delayed signal hits the reverb is active on the tail of that. Now, since I might want this beyond the outer edge of the stereo image, if the combined elements are such that hard panning to one side is not the best route, then I will pan the return 100% and create a pseudo stereo image. Otherwise I will hard pan it to the opposite side. Since I want it to engage in the mix depending on sound source quality and playback loudness level, I will adjust the volume of this slapback so that if the vocals are of high quality, I will leave the slapback on a mild volume setting, so that when you turn up the playback level louder you suddenly get that added sense of harmonic richness in the mix. (this I usually automate too) This helps to make the listeners turn up the volume which is important for mix impact. I find both delay and reverb are important, but reverb is an effect that requires more from the engineer. It happens that I delay the whole frequency range but have reverb applied only on particular bands of the delay. It depends on what effect I want the reverb to have on the mix.
The goal is to establish the quality as early as possible, so I don't accept the recording into mixing if the vocal harmony are out of tune. But once the vocal harmony is really great, I add tuning on particular spots to place it into harmonic nirvana. I find there is no reason why not to do this, it is more beautiful with tuning applied, but I find it should be added cleverly as sweet spots in the mix.
If the harmony consists of low male vocals, then a lot of focus goes into the peak and rms of the low end of it. Generally it does not sound good to have too much low frequencies panned too widely and there are frequencies in the mix that need to stand out in terms of low end without causing mix density, sound sources such as bass, kick and snare. But because low frequencies add size, I leave it in there without reverb applied so that when you turn up the volume it becomes active and causes the mix to grow. So sometimes the mute button is the best, sometimes it is the best to leave it in the mix but at very low volume. It also depends on what is in the arrangement, because you might want to extend various sound sources in the same frequency range psychoacoustically.
However, nine times out of ten, the low end of the lead vocals is muted, purely to combat dense mix density. I do like the low end of the lead vocals, but I really dislike dense mixes. Mixes can also be very bright however get a very sharp attack in the upper mids and highs. To combat this one has to ensure there is enough low frequencies on the combination of the other sound sources in the mix. So it's all put right through proper balancing decisions.
With vocal harmony it is the mid range I focus on, because the mix impact is mainly in the mid range. The focus is then on the rms and peak of the mid range, relative to the rest of the mid frequencies in the mix. This is hence the dynamic portion of how I work with the vocal harmony and it often happens that I expand rather than compress the peak, especially in the low mids. I might compress the mids (side chained) and expand the low mids. Sometimes I might not be able to do this on both speakers though or that I want a separate setting on each speaker, each speaker is its own discrete mix.
Fundamentally though, just as with any other sound source in the mix, it is the rms and peak balancing of the sound source's frequency range that is the critical part. In the pop genre you will see that the side is often very steeply cut in the lows and highs. This is mainly because in that way the size is increasing more exponentially as you turn up the playback volume (the size is towards each end of the frequency range especially on the side component) so that it becomes super good sounding at very loud volume, but it is also done to create the perception of a dynamically round low mid range that creates a very warm experience at lower playback levels, because that is very peaceful and delicate and gives this very comfortable listening experience. (= sweet sounding mix) This warmth is achieved with hardware, cannot be achieved with software. Sounds like this in the pop genre (unfortunately this mix was not properly bounced):
