Selig Leveler question

This forum is for discussing Rack Extensions. Devs are all welcome to show off their goods.
Post Reply
johnghadimi
Posts: 39
Joined: 13 Dec 2016

03 Nov 2020

I'm looking to optimize my vocal mixing workflow, and I'm hoping Selig leveler will have a place in that "chain". Before I feed the vocals into a compressor and other processes down the chain, I'd like to tackle any gain inconsistencies and also gain stage prior to getting into EQ & Compression, FX, etc..

Will Leveler technically handle the gain smoothing of the vocals, thus allowing me to then target a gain stage after (like -18dbFs)? I've never used Leveler, but looking at the documentation, it seems this could be the right tool for that. Hoping other Leveler users can confirm this. Thanks.

User avatar
guitfnky
Posts: 4408
Joined: 19 Jan 2015

03 Nov 2020

yes, it’s pretty great for that. one thing to keep in mind with it though (and I think with any upward compressor) is to be mindful of your sibilance. I would routinely add Curve/Leveler and then wonder why my vocals were getting so ridiculously bright. turned out my quietest sung notes weren’t always as loud as some of the sibilance. as a result, in the process of setting up the compression, following the recommendations, it results in the sibilance getting caught up in the upward compression.

so if it ever run into something like that, just make sure to account for it somehow (like putting any de-essing before the compression to make sure the sibilance is below the upward compression threshold).

same thing would apply for any other sounds like loud breaths, etc.
I write good music for good people

https://slowrobot.bandcamp.com/

johnghadimi
Posts: 39
Joined: 13 Dec 2016

03 Nov 2020

This is great advice. Thank you.

User avatar
selig
RE Developer
Posts: 11685
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

04 Nov 2020

johnghadimi wrote:
03 Nov 2020
I'm looking to optimize my vocal mixing workflow, and I'm hoping Selig leveler will have a place in that "chain". Before I feed the vocals into a compressor and other processes down the chain, I'd like to tackle any gain inconsistencies and also gain stage prior to getting into EQ & Compression, FX, etc..

Will Leveler technically handle the gain smoothing of the vocals, thus allowing me to then target a gain stage after (like -18dbFs)? I've never used Leveler, but looking at the documentation, it seems this could be the right tool for that. Hoping other Leveler users can confirm this. Thanks.
Short answer, yes. Long answer…it's not a magic bullet.
First, Leveller is still a compressor, though a more transparent one. Once you level a vocal, you really don't need compression, or much compression. You certainly can add further compression, but it's affect will be less obvious than if no Leveler is used. It is going to be a case by case situation where you adjust the amount (Blend) of Leveling vs the amount of compression to get the desired overall effect.
To your specific question about levels, which has nothing to do with gain staging btw (an old analog term), it's easy to hit a target peak level with Leveller. If your audio is already hitting the desired highest peak, say -18 dBFS, you simply set the Target to match. If not, after setting the target you adjust the output until the signal reaches the desired level. For example if your vocal goes over your final desired level you need to set the Leveler's Target to match the highest peak, then subtract the difference on the Output. So if the vocal peaks at -12 dBFS, set the Target there and the Output to -6 dB (12 - 6 = 18)
At this point, the Curve parameter controls how much of the audio below the Target is affected/compressed (upwards).
If for example your softest word is 20 dB below the loudest peak above, simply set Curve to 20 dB. With a Blend of 100%, all signals within this 20 dB range will be increased to match the target level, at the rate set by Recovery. With a Blend of 50% (default), you get a 2:1 compression ratio, where the signal is adjusted half way to the target. Meaning, a signal that is 10 dB below the target will be raised by half that, or 5 dB.
Note, the Target and Curve settings are going to strictly depend on the audio level coming into the Leveler (so make it first in your vocal chain!), while the Recovery and Blend (and Output) settings are more up to you to choose depending on how much you want to level the vocal (or whatever you're using Leveler for). Make sense?
Selig Audio, LLC

User avatar
guitfnky
Posts: 4408
Joined: 19 Jan 2015

04 Nov 2020

selig wrote:
04 Nov 2020
johnghadimi wrote:
03 Nov 2020
I'm looking to optimize my vocal mixing workflow, and I'm hoping Selig leveler will have a place in that "chain". Before I feed the vocals into a compressor and other processes down the chain, I'd like to tackle any gain inconsistencies and also gain stage prior to getting into EQ & Compression, FX, etc..

Will Leveler technically handle the gain smoothing of the vocals, thus allowing me to then target a gain stage after (like -18dbFs)? I've never used Leveler, but looking at the documentation, it seems this could be the right tool for that. Hoping other Leveler users can confirm this. Thanks.
Note, the Target and Curve settings are going to strictly depend on the audio level coming into the Leveler (so make it first in your vocal chain!)
unless you need to de-ess.
I write good music for good people

https://slowrobot.bandcamp.com/

User avatar
selig
RE Developer
Posts: 11685
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

04 Nov 2020

guitfnky wrote:
03 Nov 2020
yes, it’s pretty great for that. one thing to keep in mind with it though (and I think with any upward compressor) is to be mindful of your sibilance. I would routinely add Curve/Leveler and then wonder why my vocals were getting so ridiculously bright. turned out my quietest sung notes weren’t always as loud as some of the sibilance. as a result, in the process of setting up the compression, following the recommendations, it results in the sibilance getting caught up in the upward compression.

so if it ever run into something like that, just make sure to account for it somehow (like putting any de-essing before the compression to make sure the sibilance is below the upward compression threshold).

same thing would apply for any other sounds like loud breaths, etc.
I find sibilance issues more often worse with downwards compression than upwards/Leveler as a general rule.

If using Selig DeEsser you can put it anywhere in the chain, it won't matter. It all depends on how loud the sibilance/breaths/etc are compared to the main voice. In most cases the sibilance and breaths will fall below the loudest vocal so it's easy to set the Curve setting to avoid "lifting" those things too much. But if you've compressed the vocal a lot when recording, that may not be easy to do. One fairly simple trick for any dynamics processing "problem" like this is to automate the threshold or input to the compressor for the handful of places where the background is overly affected. This is also common when trying to get a gate to ignore a few rogue hits on a drum kit.

I generally hesitate putting things in front of the Leveler because changes there will mean likely adjustments to the Leveler - so if you don't want to be making changes in multiple places, better to put Leveler first so you can move on to focus on other things!
Selig Audio, LLC

User avatar
guitfnky
Posts: 4408
Joined: 19 Jan 2015

04 Nov 2020

selig wrote:
04 Nov 2020
guitfnky wrote:
03 Nov 2020
yes, it’s pretty great for that. one thing to keep in mind with it though (and I think with any upward compressor) is to be mindful of your sibilance. I would routinely add Curve/Leveler and then wonder why my vocals were getting so ridiculously bright. turned out my quietest sung notes weren’t always as loud as some of the sibilance. as a result, in the process of setting up the compression, following the recommendations, it results in the sibilance getting caught up in the upward compression.

so if it ever run into something like that, just make sure to account for it somehow (like putting any de-essing before the compression to make sure the sibilance is below the upward compression threshold).

same thing would apply for any other sounds like loud breaths, etc.
I find sibilance issues more often worse with downwards compression than upwards/Leveler as a general rule.

If using Selig DeEsser you can put it anywhere in the chain, it won't matter. It all depends on how loud the sibilance/breaths/etc are compared to the main voice. In most cases the sibilance and breaths will fall below the loudest vocal so it's easy to set the Curve setting to avoid "lifting" those things too much. But if you've compressed the vocal a lot when recording, that may not be easy to do. One fairly simple trick for any dynamics processing "problem" like this is to automate the threshold or input to the compressor for the handful of places where the background is overly affected. This is also common when trying to get a gate to ignore a few rogue hits on a drum kit.

I generally hesitate putting things in front of the Leveler because changes there will mean likely adjustments to the Leveler - so if you don't want to be making changes in multiple places, better to put Leveler first so you can move on to focus on other things!
agreed, generally. I tend to have louder sibilants in my vocals, even before any processing (I never compress on the way in)—for me, it’s not just a few places, it’s fairly consistent, so automation isn’t a good option, which is why I say you may want to put a de-esser ahead of Leveler IF you have a similar issue (which many likely won’t—but it’s still worth being aware of).

it was something I wish I’d realized much sooner because it resulted in me scratching my head and avoiding using Leveler for a long time, despite thinking it’s a great tool.
I write good music for good people

https://slowrobot.bandcamp.com/

johnghadimi
Posts: 39
Joined: 13 Dec 2016

04 Nov 2020

Thank you for the response. Totally makes sense. I did a quick experiment last night and the results were great. :) I had it first in the chain, and due to some noticeable sibilance presence, I added a de-esser after it to fix the issue. I typically have a gain plugin as the very first thing in the chain to gain stage, so as far as the chain goes, this seems to do the trick for me: gain stage plugin -> Leveler -> De-ess
selig wrote:
04 Nov 2020
johnghadimi wrote:
03 Nov 2020
I'm looking to optimize my vocal mixing workflow, and I'm hoping Selig leveler will have a place in that "chain". Before I feed the vocals into a compressor and other processes down the chain, I'd like to tackle any gain inconsistencies and also gain stage prior to getting into EQ & Compression, FX, etc..

Will Leveler technically handle the gain smoothing of the vocals, thus allowing me to then target a gain stage after (like -18dbFs)? I've never used Leveler, but looking at the documentation, it seems this could be the right tool for that. Hoping other Leveler users can confirm this. Thanks.
Short answer, yes. Long answer…it's not a magic bullet.
First, Leveller is still a compressor, though a more transparent one. Once you level a vocal, you really don't need compression, or much compression. You certainly can add further compression, but it's affect will be less obvious than if no Leveler is used. It is going to be a case by case situation where you adjust the amount (Blend) of Leveling vs the amount of compression to get the desired overall effect.
To your specific question about levels, which has nothing to do with gain staging btw (an old analog term), it's easy to hit a target peak level with Leveller. If your audio is already hitting the desired highest peak, say -18 dBFS, you simply set the Target to match. If not, after setting the target you adjust the output until the signal reaches the desired level. For example if your vocal goes over your final desired level you need to set the Leveler's Target to match the highest peak, then subtract the difference on the Output. So if the vocal peaks at -12 dBFS, set the Target there and the Output to -6 dB (12 - 6 = 18)
At this point, the Curve parameter controls how much of the audio below the Target is affected/compressed (upwards).
If for example your softest word is 20 dB below the loudest peak above, simply set Curve to 20 dB. With a Blend of 100%, all signals within this 20 dB range will be increased to match the target level, at the rate set by Recovery. With a Blend of 50% (default), you get a 2:1 compression ratio, where the signal is adjusted half way to the target. Meaning, a signal that is 10 dB below the target will be raised by half that, or 5 dB.
Note, the Target and Curve settings are going to strictly depend on the audio level coming into the Leveler (so make it first in your vocal chain!), while the Recovery and Blend (and Output) settings are more up to you to choose depending on how much you want to level the vocal (or whatever you're using Leveler for). Make sense?

User avatar
raveled
Competition Winner
Posts: 85
Joined: 08 Jul 2016

19 Nov 2020

I thought I’d share something I just tried. Turns out, Leveler is an excellent “sustainer”. While figuring this out of course is no rocket science, I just had not thought about it that way before.

So let’s say you want to add some extra sustain on a long note of a lead guitar. First, set up Leveler as usual. Then automate the curve fader and let it slide gently to the right on that particular note. Finally, when the note has reached its end, let the slider go left and return to the initial state.

To me, this worked amazingly well, and offers an alternative to e.g. automating a gain knob. And since the Leveler is already there first in my chain, it’s good for rack economics. Lastly, the wet/dry parameter can be automated to further manipulate the behavior of the tail.

Post Reply
  • Information
  • Who is online

    Users browsing this forum: No registered users and 28 guests