Newbie: Mixing a highly dynamic singer

Air For Wings

New member
K. I'm a singer with tech background but I am not a recording engineer. My main mixing problem is that I'm highly dynamic with a broad vocal resonance (contralto voice) and don't know how to mix for that and still keep presence without distortion.

(BTW I used to do tech in radio and in TV so I'm the worst kind of newbie. The kind that imagines they can take this on! LOL )

I'm recording and mixing my own songs using Ableton.
I've got my studio set up to get good clean sound for the soft, intimate vocals and the power notes in the same session. No noticeable background noise. (When I record my vocals, it's with my pre-mixed backing track in a headset, so just my voice.)

When I go to mix:
I've been importing and then consolidating and normalizing the vocal track.
There's often a 9db or even 12 db difference between the softer parts and the power notes on most of my vocal tracks. (That's with me backing off the mic on the power notes.)

If I keep the levels down, to accommodate the power notes, the soft parts get drowned by the music or, alternately, have no presence.
If I crank up the soft parts, then I overdrive the power notes. Applying a compression filter / plug in has not been fixing that. The power notes still sound distorted.

If I do different sound levels for the quiet parts vs the power notes, they end up having different "presence", which weakens the song and also gets distracting.

If I notch the background music to pop the vocals at the tricky parts, the background music almost entirely drops out because my vocals have a broad resonance on them, so I have to notch too much of the background, (and otherwise the notching doesn't work).

My voice is also contralto, which gives my voice its unique sound. I don't want to lose that in the mixing. Although, on the power notes, I understand one can successfully compromise.

I do not know how to mix this.

If you folks can help me figure this out, there's a huge back load of music I can finish that's going to come tumbling out. That would be joy!
 
When I go to mix:
I've been importing and then consolidating and normalizing the vocal track.
Why? (Where the notion that being anywhere near a first step in mixing started beats me.

There's often a 9db or even 12 db difference between the softer parts and the power notes on most of my vocal tracks. (That's with me backing off the mic on the power notes.)
There's a good first start.

If I keep the levels down, to accommodate the power notes, the soft parts get drowned by the music or, alternately, have no presence.
If I crank up the soft parts, then I overdrive the power notes. Applying a compression filter / plug in has not been fixing that. The power notes still sound distorted.

Now we start mixing. And where we begin to address things.
1) One, levels. -Track automation (AKA... 'mixing' :>)
2) Addressing inconsistent tones. Item one may be 'why the voice has to loose edge or focus partly can be addressed in the singing but.. Addressing it with eq, again here's where automation can come in. Some folks like to split sections into two tracks. (Not a big fan of that - track 'shifts/transitions usually are nowhere as simple as that and... we can and often do.. automate a lot more than levels'.

If I do different sound levels for the quiet parts vs the power notes, they end up having different "presence", which weakens the song and also gets distracting.
Same as above.

If I notch the background music to pop the vocals at the tricky parts, the background music almost entirely drops out because my vocals have a broad resonance on them, so I have to notch too much of the background, (and otherwise the notching doesn't work).
Try being a little more subtle addressing things 'mixing :>) Fixing -leveling, eq, compression etc- the vocal first makes 'sit in the mix easier. Then move on to building 'a mix around the voc.

My voice is also contralto, which gives my voice its unique sound. I don't want to lose that in the mixing. Although, on the power notes, I understand one can successfully compromise.
Good, don't do that. As a mix progresses you move on to more and more finer points. The previous work makes it easier to see where and what to do next.

If you folks can help me figure this out, there's a huge back load of music I can finish that's going to come tumbling out. That would be joy!
No problem.
Joy!
And welcome to the hunt :listeningmusic:
 
Last edited:
I suspect there are two small problems in the same department - first you've probably been preoccupied with preventing the input stage distorting on the peaks, and then the quieter sections get a bit lost, and bringing up their level brings background graininess with it, then trying to make the huge dynamic range fit in the limited range of a song.

Although normally I'm an advocate of recording with no processing, in this case, I'd gently compress what was coming in, and then gently compress again before I start using the channel fader in the mix. The fader range will be more manageable, but the compressor won't start pumping which is always horrible, but common with ultra dynamic singers. I've also sometimes moved to a dynamic mic with no proximity effect - like the EV RE20 and RE320 - and then letting the singer use mic technique to pull off on the belts and move in for the gentler parts.
 
I've also sometimes moved to a dynamic mic with no proximity effect - like the EV RE20 and RE320 - and then letting the singer use mic technique to pull off on the belts and move in for the gentler parts.
I've been using a Behringer C-1, through a Scarlette mix control.
I've ordered at Sennheiser e945 Supercardioid Dynamic. Hoping to see a difference.

Backing off the mic is working a bit but at extremes, just reduces presence.

BTW Someone suggested double mic-ing. I haven't been able to make that work successfully yet. Mic's seem to each have their own imprint and then I end up having to adjust for that, so just shifting the problem to a different arena.

in this case, I'd gently compress what was coming in, and then gently compress again before I start using the channel fader in the mix. The fader range will be more manageable, but the compressor won't start pumping which is always horrible, but common with ultra dynamic singers.

Will that lose tonal quality?
 
OK, Mixit, if I follow you..
Some suggest I split the tracks and handle the power notes on separate tracks giving them separate treatments; but you're not a fan of that. That track splitting presents problems later?
I've had someone suggest using the Ableton "Multiband Dynamics" plug in to do selective high pass filters that could kick in only on the power notes.

Does anyone know if this could work?
 
Not familiar with Ableton but the problem is not terribly unusual. Can we assume you are working with some sort of volume automation first and then sending the track to compression? Careful automation can sometimes get you close enough for further processing to work in cases like this.
 
Basically, I've been importing the vocals and then setting them so the power notes are short of peak. With my new mic set up, I don't have to adjust much to do that.
 
You folks are being wonderfully responsive and helpful.
While I see what other ideas you might have, I think what I'm going to try is the advise on using the Multiband Dynamics plug in which gives selective frequency high pass and applying that only to the power notes. That should keep me busy for a couple of weeks while I figure out how to apply that. So any other, easier to apply ideas are welcome!
 
If you are recording the voice alone then you can adjust the takes before mixing so the problem wont happen.

If your tracks are at different volume levels them tweak them separately.

If one track goes low and also loud then add compression to fix that. Or possibly use your DAW to adjust just the low and high parts levels separately.

Or did I miss something else that is really the problem?
 
One technique that I've been using the last few years to get more consistent presence in a mix:
have 1 untreated version of the vocal in the mix.
Send that to a bus with a ton of compress. Squash it flat and boring.

Mix between those two versions throughout the song.
 
Track twice - once for the soft stuff, once for the loud stuff (with the gains adjusted accordingly) then comp the parts together.
 
Back
Top