Get control of vocals in the mix
This is a technique to control the vocal placement in the mix. Make it stand out or blend. The principle is to group instruments separately and vocals separately to make room for or reduce vocal characteristics with EQs.
Start by routing each and everyone of the instruments to a bus (BUS1). Now, by lovering the bus volume you should only hear the vocals.
If yo have more then one vocal track you should group them to another bus (BUS2).
BUS1 BUS2 | | Instruments vocals
The example in this blog post has only one vocal track, so there’s no actual benefit of routing it to a bus.
INSTUMENTBUS | | Instruments Lead Vocals
On vocal track ( or vocal bus if created):
Use a parametric equalizer and sweep over the frequency-band to find the vocal characteristics. When found, gain a few dB. Then isolate it by narrowing the bandwidth.
On instrument bus:
Since the instrument bus contains several instruments, it’s difficult to identify a joint characteristic. Therefore a simple approach is to use a parametric equalizer with the same center frequency as for the vocal track. Reduce gain a few dB. Then isolate it by narrowing the bandwidth.
What you have managed now are the vocals to stand out. Now lets compare the original and the EQed song.
Original song
EQed song
If you instead want the vocals to step back or blend. Invert the gain of the vocals and instruments.
Tags: vocals
September 20th, 2010 at 10:50 am
Hi There !
Thx for this nice and simple tutorial. I’ll try that for sure.
Still I have a few questions.
* Why does the second sample song sound louder ? Is it because the use of this technique allows to raise the overall volume without risking clipping in these frequencies ?
* Do the parameters of the 2 equalizers need to be a perfect match (meaning the same values) or are they chosen by a try/error method (and thus can be different values) ?
Thanks a lot for your answers
jy (Paris, France)
PS: I’ve just discovered your blog and will be following it from now on !
September 21st, 2010 at 12:27 am
Hi, thx and welcome!
I would say it allows avoiding risk of clipping in any frequency. Because you choose the frequency not by peak but where the vocal characteristics are most prominent. If you replace the vocals in the example with a flute, the characteristics would be the frequencies where it sounds most like a flute and not any other instrument.
The eqs don’t need to be perfect match but it’s a simple approach.
If you can find conflict frecuencies between vocals and instruments with 2nd eq then use them. But it can be difficult to isolate throughout a whole song with many instruments on the same bus.
May 6th, 2011 at 4:47 pm
Nice topic – respect !
June 20th, 2011 at 8:06 pm
so you want to learn how to hack a facebook account
August 24th, 2011 at 3:18 pm
@Winston Three things are certain: Death, taxes, and lost data. Guess which has occurred
May 4th, 2012 at 3:54 pm
[...] unas semanas como borrador, me he decidido a traducir esta clásica entrada de LinuxAudioBlog para que no caiga en el [...]
July 15th, 2012 at 6:03 am
Nice topic – respect !