Trouble getting vocal harmonies to blend...HELP!

[guitarsg1]

Hi I am getting a rather decent vocal sound out of these 2 guys who are decent singers. I am using a Studio Projects C1 mic and the music is like psychadelic rock (Floydish).
The song has 2 main vocal lines that are melodicly different from one another, but are harmonized. My problem is that I am having trouble getting the harmonies to be distinguishable-yet-well-blended. Like I said, the 2 vocal tracks sound pretty good each on thier own, and are each in key/ the harmonies match up correctly, but when they are together they sound real muddy. Also, when they sing this song w.out recording there in my room w/ and acoustic guitar, the vocals/harmonies sound great.

I remember about 2 years ago when a past band I was in recorded in a semi-pro studio, and when the vocal harmonies were done w/ our not-so great singer, it sounded like an angelic choir, no joke.

So, my question is, does anyone have some EQ/reverb tips or techniques they use to make harmonies sound smoother and blended nice? A sound I am kinda going for is like Pink Floyds harmonies and/or weezer's harmonies, particularly on the green album. They are so smooth, it sounds like 1 voice singing the 2 different harmonies at physically the same time. Anyone? Thanks

 

[nessbass]

The major difference between the case you're working on and the examples you mentioned is that you are now working with two different singers and voices, whereas in your examples all parts were sung by one singer. And that helps a lot, because on voice singing two harmonic parts sound like, well, one voice singing two parts!

One thing you could try however, is to compress them together and then send them to the same reverb. I don't know if one could be successful trying to make the two voices less dissimilar by the use of EQ, but unless you have to pay for your studio time, I would give it a try. If one of the two parts is more of a backing track with one being the main vocal, try to eq the backing track. But maybe this is all a bucket full of shit, and it's best to leave the EQ alone. But compressing the two together should make a difference, IMHO.

David.

 

[Track Rat]

Ditto on the compressor. I wouldn't buss them through the same compressor though, just try to level each one individualy and the mix to taste. Something else you might try if you have the tracks is to have them double or even quadruple track there parts for that angelic sound (think Pet Sounds/ Beach Boys). Just an idea.

 

[Rev E]

You may want to try Eqing out some of the lows. You can use a low shelf EQ or a parametric with a low Q (~ < 1) and reduce frequencies below 100 - 200. This may clean up the tracks a bit. These are just ballpark figures, all of this depends on what the recording sounds like.

Another thing that you can do if one of the tracks is a background track is to EQ out the mid-range where the voice is strong to the human ear. Try EQing out (with a notch EQ = high Q ~8-10) some of the vocal at around 1 kHz - 1.5 kHz. This will shift the vocals a bit back from the lead vocal. In order to get the vocals to blend, you can notch out slightly different places for each harmony. For instance, for the lower harmony you may notch out a few dBs at 1.5 kHz and with the higher harmony you notch out 1 kHz... the rationale being that you're taking out some of the higher voice mids on the "lower" vocal and leaving those frequencies intact on the higher harmony, while taking out some of the lower mid frequencies on the lower vocal. OR you could switch it the other way... this is just a ballpark figure AND more importantly, just an "approach" to blending the vocals.

Regarding reverb and other spatial effects, if you listen to commercial mixes of similar styles (or any style for that matter), you'll notice that if you have two vocals that are competing lead lines (or "call-and-answer" type vocals), one of them will remain relatively natural (a bit of tasteful EQ, some verb and/or a touch of delay). The other line will be more noticeably effected (a drastic chorus/reverb, a "telephone" EQ/distortion effect, a mild-but-noticeable distortion effect, a flanger, etc..... So if the two vocal lines are truly lead lines make a decision as to which one will remain "natural" and put more drastic effects on the other one.

Rev E

 

[TonyDawgz]

For an easy fix, all you need to do is go -5db @ 5K on the "weaker" singer. Do it to each one seperately then whichever sounds better, leave it that way. This is for any muddy vocals you have (like choruses or multiple talkers). You may also need to raise @5K for the "lead"

Hope this helps

 

[Rev E]

I respectfully disagree with the "5K" approach. At 5K, the voice generally has the strongest "s"'s. 5k is the frequence that a desser will be used on. Increasing this area would likely overload your "s" sound even more. Plus, the C1 is a bright mic at 5K (I have one also). Voices have the most information in the mids. Focus on EQing in that area. 5k would more aptly qualify as "highs"... Blending vocals is usually more about lows and mids than highs.

Guitarsg1, why don't you post a link of the two vocals as MP3's and let us play around with the tracks for ourselves. Hearing is worth a whole lot more in the audio world than theory.

Rev E