I heart Selig Leveler

This forum is for discussing Rack Extensions. Devs are all welcome to show off their goods.
User avatar
dvdrtldg
Posts: 2386
Joined: 17 Jan 2015

27 Oct 2017

I do a lot of radio production in WaveLab, and constantly have to level out (spoken) voice tracks. Selig Leveler is the tool for the job, but it's always bothered me slightly to have to drag the audio into Reason, apply leveling, and export back out to WaveLab

So today I thought I'd give Waves Vocal Rider a try. It's cool, says everyone, because it's not a compressor and therefore avoids the problems that can sometimes occur with compression

Except that with Selig Leveler, there are no problems. It really is amazingly seamless and transparent. I think I'd rather taken Leveler for granted until I put it up against this "dedicated" vocal rider, it absolutely smokes the Waves product

Not sure if this really deserves a dedicated thread, but I'm on a bit of a Reason high today

User avatar
Zac
Posts: 1784
Joined: 19 May 2016
Contact:

27 Oct 2017

Nothing wrong with a bit of enthusiasm. I totally agree, it's an amazing plug in. I rarely use vocals but I still get so much use out of it.

User avatar
jayhosking
Posts: 613
Joined: 28 Nov 2016
Contact:

27 Oct 2017

Maybe it's finally time to give it a shot; I love these sort of testimonials and with the new version of Reason, I'm in that honeymoon phase again. Does it ever go on sale? I know it's part of the mix/master bundle, but I own much of that already and so it wouldn't be a good value to me even at the discount.

User avatar
mbfrancis
Posts: 647
Joined: 02 Feb 2015
Location: Orange County, CA
Contact:

27 Oct 2017

I recently used it on a violin track where she kept dropping out, turning away from the mic. Worked great without squashing any peaks.
Producer, songwriter, multi-instrumentalist. I make indie pop as Port Streets, 90s/shoegaze as Swooner, and Electro as Yours Mine.

User avatar
Eagleizer
Posts: 102
Joined: 15 Jan 2015
Location: Thailand

27 Oct 2017

Yep! I use it in every project. Awesome device! It is also great for
removing background noise.

Cheers :)

User avatar
Reasonable man
Posts: 589
Joined: 14 Jul 2016

27 Oct 2017

It's damm expensive . The example on you tube though listens to it 'level' out vocals. I personally hate the way vocals to certain genres are mixed . Auto-tuned to un-natural/false state with zero dynamics making it again seem unnatural and unrealistic. Leveler does the latter, it squeezes the dynamic range out of vocals (i appreciate that this is the style of alot of modern genres) . On synths and elaborate song tranistions though i can imagine it being the dogs bollox

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

27 Oct 2017

Reasonable man wrote:It's damm expensive . The example on you tube though listens to it 'level' out vocals. I personally hate the way vocals to certain genres are mixed . Auto-tuned to un-natural/false state with zero dynamics making it again seem unnatural and unrealistic. Leveler does the latter, it squeezes the dynamic range out of vocals (i appreciate that this is the style of alot of modern genres) . On synths and elaborate song tranistions though i can imagine it being the dogs bollox
Hi, Giles from Selig Audio here! Always excited to see anyone who appreciates our products, and willing to discuss issues with those who don’t!

“Expensive” is relative. I tried to price it as reasonably as possible and give it as much power and as high a quality as possible.

Leveler IS able to squeeze all the dynamics out of a signal (and do it amazingly transparently, FWIW).

However, I don’t personally know anyone who uses it that way on vocals, as IMO it doesn’t sound at all natural. Leveler has a Blend control that allows you to use from zero to 100% of it’s effect. I hardly ever use more than 50% and very often use 20-30%. Hopefully I’ve explained the power of that control in my videos and the User Guide.

Bottom line…
The user has total control over how much leveling to apply, so if you choose to squeeze the life out of a vocal that’s on you. Same with Auto Tune, which you can use to nudge an occasional note closer to being in tune or you can create an almost vocoder-like effect with it. Another example - our Selig DeEsser can TOTALLY take all sibilance out of a vocal - IF YOU WANT. It’s useful for parallel channels or reverb/delay sends, in case you’re wondering why that power exists!

I’ve always appreciated tools that have a wide range of available power, and have strived to provide those tools as an RE developer. While I try to guide users as to how to get the best from these tools, their use is not under my control - and my goal was not to provide tools to ‘ruin’ music production for anyone in any way! ;)


Sent from some crappy device using Tapatalk
Selig Audio, LLC

User avatar
unisyn
Posts: 45
Joined: 12 Aug 2016

27 Oct 2017

Second this.
There are many other dynamics riding plugins, but this one is extremely easy to tune.

User avatar
Reasonable man
Posts: 589
Joined: 14 Jul 2016

27 Oct 2017

selig wrote:
27 Oct 2017
Reasonable man wrote:It's damm expensive . The example on you tube though listens to it 'level' out vocals. I personally hate the way vocals to certain genres are mixed . Auto-tuned to un-natural/false state with zero dynamics making it again seem unnatural and unrealistic. Leveler does the latter, it squeezes the dynamic range out of vocals (i appreciate that this is the style of alot of modern genres) . On synths and elaborate song tranistions though i can imagine it being the dogs bollox
Hi, Giles from Selig Audio here! Always excited to see anyone who appreciates our products, and willing to discuss issues with those who don’t!

“Expensive” is relative. I tried to price it as reasonably as possible and give it as much power and as high a quality as possible.

Leveler IS able to squeeze all the dynamics out of a signal (and do it amazingly transparently, FWIW).

However, I don’t personally know anyone who uses it that way on vocals, as IMO it doesn’t sound at all natural. Leveler has a Blend control that allows you to use from zero to 100% of it’s effect. I hardly ever use more than 50% and very often use 20-30%. Hopefully I’ve explained the power of that control in my videos and the User Guide.

Bottom line…
The user has total control over how much leveling to apply, so if you choose to squeeze the life out of a vocal that’s on you. Same with Auto Tune, which you can use to nudge an occasional note closer to being in tune or you can create an almost vocoder-like effect with it. Another example - our Selig DeEsser can TOTALLY take all sibilance out of a vocal - IF YOU WANT. It’s useful for parallel channels or reverb/delay sends, in case you’re wondering why that power exists!

I’ve always appreciated tools that have a wide range of available power, and have strived to provide those tools as an RE developer. While I try to guide users as to how to get the best from these tools, their use is not under my control - and my goal was not to provide tools to ‘ruin’ music production for anyone in any way! ;)


Sent from some crappy device using Tapatalk
Shit .... sorry i didn't know you were that 'Selig' . Wasn't critising your product dude ! Was just making a commentary on modern music i guess and how i'm conviced that people will look back on alot of 'music' produced from this decade like they do with the more extreme examples of the 80's and wonder what people were smoking. If there is a few quid off on black friday i will be buying this ....obviously not for vocals! but as i said its a life saver for ironing out problem arrangements and transition levels. Its a top product for that fair play

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

27 Oct 2017

Reasonable man wrote:
27 Oct 2017
selig wrote:
27 Oct 2017


Hi, Giles from Selig Audio here! Always excited to see anyone who appreciates our products, and willing to discuss issues with those who don’t!

“Expensive” is relative. I tried to price it as reasonably as possible and give it as much power and as high a quality as possible.

Leveler IS able to squeeze all the dynamics out of a signal (and do it amazingly transparently, FWIW).

However, I don’t personally know anyone who uses it that way on vocals, as IMO it doesn’t sound at all natural. Leveler has a Blend control that allows you to use from zero to 100% of it’s effect. I hardly ever use more than 50% and very often use 20-30%. Hopefully I’ve explained the power of that control in my videos and the User Guide.

Bottom line…
The user has total control over how much leveling to apply, so if you choose to squeeze the life out of a vocal that’s on you. Same with Auto Tune, which you can use to nudge an occasional note closer to being in tune or you can create an almost vocoder-like effect with it. Another example - our Selig DeEsser can TOTALLY take all sibilance out of a vocal - IF YOU WANT. It’s useful for parallel channels or reverb/delay sends, in case you’re wondering why that power exists!

I’ve always appreciated tools that have a wide range of available power, and have strived to provide those tools as an RE developer. While I try to guide users as to how to get the best from these tools, their use is not under my control - and my goal was not to provide tools to ‘ruin’ music production for anyone in any way! ;)


Sent from some crappy device using Tapatalk
Shit .... sorry i didn't know you were that 'Selig' . Wasn't critising your product dude ! Was just making a commentary on modern music i guess and how i'm conviced that people will look back on alot of 'music' produced from this decade like they do with the more extreme examples of the 80's and wonder what people were smoking. If there is a few quid off on black friday i will be buying this ....obviously not for vocals! but as i said its a life saver for ironing out problem arrangements and transition levels. Its a top product for that fair play
Sorry if I came off sounding like I thought you were criticizing - I only wanted to be clear that it's up to the user as to how to use this (or any) device. Just because a knob goes up to 10 (or 11 for that matter) doesn't mean you have to use it there! ;)

As for the price, Selig Audio does a 50% off sale at least once a year, either in May or November (usually to coincide with the Props sales).
:)
Selig Audio, LLC

User avatar
Zac
Posts: 1784
Joined: 19 May 2016
Contact:

27 Oct 2017

Reasonable man wrote:
27 Oct 2017
selig wrote:
27 Oct 2017


Hi, Giles from Selig Audio here! Always excited to see anyone who appreciates our products, and willing to discuss issues with those who don’t!

“Expensive” is relative. I tried to price it as reasonably as possible and give it as much power and as high a quality as possible.

Leveler IS able to squeeze all the dynamics out of a signal (and do it amazingly transparently, FWIW).

However, I don’t personally know anyone who uses it that way on vocals, as IMO it doesn’t sound at all natural. Leveler has a Blend control that allows you to use from zero to 100% of it’s effect. I hardly ever use more than 50% and very often use 20-30%. Hopefully I’ve explained the power of that control in my videos and the User Guide.

Bottom line…
The user has total control over how much leveling to apply, so if you choose to squeeze the life out of a vocal that’s on you. Same with Auto Tune, which you can use to nudge an occasional note closer to being in tune or you can create an almost vocoder-like effect with it. Another example - our Selig DeEsser can TOTALLY take all sibilance out of a vocal - IF YOU WANT. It’s useful for parallel channels or reverb/delay sends, in case you’re wondering why that power exists!

I’ve always appreciated tools that have a wide range of available power, and have strived to provide those tools as an RE developer. While I try to guide users as to how to get the best from these tools, their use is not under my control - and my goal was not to provide tools to ‘ruin’ music production for anyone in any way! ;)


Sent from some crappy device using Tapatalk
Shit .... sorry i didn't know you were that 'Selig' . Wasn't critising your product dude ! Was just making a commentary on modern music i guess and how i'm conviced that people will look back on alot of 'music' produced from this decade like they do with the more extreme examples of the 80's and wonder what people were smoking. If there is a few quid off on black friday i will be buying this ....obviously not for vocals! but as i said its a life saver for ironing out problem arrangements and transition levels. Its a top product for that fair play
The 'obviously not for vocals' thing I would definitely disagree with. I think it's superb for vocals. I've been using this plug in for a long time now and although Selig advises using 50% wet or less I think there is many more important considerations.

For me the most important is the db depth you want it to look for to 'level'. For me it is quite often less than 6db, sometimes even less than 3db. I don't want it reaching down to the noise flour to bring up that much. Most often I just wish an instrument or vocall has more 'volume consistency' and this does it so well once you understand it and what it is doing.

If you reach down to 16db to bring up those levels, of course it will sound unnatural..
A breath of air as loud as a pronounced word... yuk. Used wisely I really think it's the most innovative RE out there that isn't a 'creative fx'.

EdGrip
Posts: 2343
Joined: 03 Jun 2016

27 Oct 2017

Sounds like it could be worth doing a VST version, perhaps?

User avatar
qn5
Posts: 137
Joined: 29 Nov 2016
Location: NYC
Contact:

27 Oct 2017

I use this daily. I really do wish there was an "auto" button or a default mode that automatically dialed in the ideal curve based on the highest/lowest peak on the track you're applying it too. It's a bit of a pain to have to dial it in/do the math on every instance of it. Ver.2 perhaps? :)
QN5 MUSIC
qn5.com

WongoTheSane
Moderator
Posts: 1851
Joined: 14 Sep 2015
Location: Paris, France

27 Oct 2017

Protip: Level your samples before importing into Grain. It does wonder in getting rid of volume fluctuations on most material. Drop the sample into the sequencer, drag a Leveler in the channel's FX slot, play the sample and set the Target accordingly, then bounce the sample and import into Grain. Done!

User avatar
Ottostrom
Posts: 845
Joined: 13 May 2016

28 Oct 2017

Zac wrote:
27 Oct 2017
For me the most important is the db depth you want it to look for to 'level'. For me it is quite often less than 6db, sometimes even less than 3db. I don't want it reaching down to the noise flour to bring up that much. Most often I just wish an instrument or vocall has more 'volume consistency' and this does it so well once you understand it and what it is doing.

If you reach down to 16db to bring up those levels, of course it will sound unnatural..
A breath of air as loud as a pronounced word... yuk. Used wisely I really think it's the most innovative RE out there that isn't a 'creative fx'.
I have always struggled with the Leveler and I think this is the reason why. Thanks for the great tip Zac!
Also, what is your preferred recovery setting when dealing with vocals?

User avatar
EnochLight
Moderator
Posts: 8405
Joined: 17 Jan 2015
Location: Imladris

28 Oct 2017

If I had any feature request for Selig Leveler, it would be to have sample loading for analyzation, so I wouldn't have to play the whole track just for it to detect the lowest level. Not sure how that could be implemented, though. Otherwise, that thing is near perfect.
Win 10 | Ableton Live 11 Suite |  Reason 12 | i7 3770k @ 3.5 Ghz | 16 GB RAM | RME Babyface Pro | Akai MPC Live 2 & Akai Force | Roland System 8, MX1, TB3 | Dreadbox Typhon | Korg Minilogue XD

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

28 Oct 2017

Ottostrom wrote:
28 Oct 2017
Zac wrote:
27 Oct 2017
For me the most important is the db depth you want it to look for to 'level'. For me it is quite often less than 6db, sometimes even less than 3db. I don't want it reaching down to the noise flour to bring up that much. Most often I just wish an instrument or vocall has more 'volume consistency' and this does it so well once you understand it and what it is doing.

If you reach down to 16db to bring up those levels, of course it will sound unnatural..
A breath of air as loud as a pronounced word... yuk. Used wisely I really think it's the most innovative RE out there that isn't a 'creative fx'.
I have always struggled with the Leveler and I think this is the reason why. Thanks for the great tip Zac!
Also, what is your preferred recovery setting when dealing with vocals?
If it's working for you, great…but for others…
I disagree, at least in my experience - if you reach down to 16 dB AND use 100% Blend, it probably WILL sound unnatural - but that's not how Leveler was designed to work (certainly with vocals)! I regularly use that much "depth" and get an incredibly natural sound because I only use as much Blend as it takes to get the job done - often 50% or less.

As for noise floor, it's typically 50-60 dB or more below the signal, so reaching down 16 dB should never bring up the noise floor - this is one of the 'features' of the way Leveler works as opposed to other upwards (or downwards for that matter) compressors!

As a general rule, especially with vocals or similar, you'll not get volume consistency unless you set the Curve to include the entire signal dynamic range. So if you set Curve to 3-6 dB, then everything below that will not be affected. If you have a signal that has more than 3-6 dB dynamic range, you will "warp" the dynamics, squishing the top few dB up to the top and leaving the rest down below - effectively creating an EXPANSION rather than COMPRESSION result (but not exactly either one).

So the "key" to being subtle is to use lower BLEND amounts. Think of Curve as simply defining the area you want to affect, and Blend as defining how much affect you want to apply to the output. Blend is VERY much like Ratio in a downwards compressor, such that a setting of 50% Blend is equal to a Ratio of 2:1. Curve, combined with Target, is like the Threshold in a downwards compressor. It's important to remember the results will be TOTALLY different with the Leveler, that is to say the comparisons end there!

So if you want to "level" a performance that covers 15 dB (random number for this example), you MUST set the Curve and Target accordingly. From there you dial in just as much Blend as it takes to bring up the lower parts (when in doubt, use less Blend).
Selig Audio, LLC

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

28 Oct 2017

EnochLight wrote:
28 Oct 2017
If I had any feature request for Selig Leveler, it would be to have sample loading for analyzation, so I wouldn't have to play the whole track just for it to detect the lowest level. Not sure how that could be implemented, though. Otherwise, that thing is near perfect.
Once you load a sample, how do you find the lowest level? Technically, the lowest level is noise. The only way to identify the lowest level is to decide where YOU want the lowest level to be. It's not possible to use metering to find this "musical" lowest level, as there are too many variables. It's really a subjective thing.

So unless you know of some existing technology that can detect the softest note/word or a recording, this isn't possible - it's the same probable that prevents us from having a button that does this in real time, so speeding up the process won't make it suddenly possible to do so. Hopefully I'm making sense…

BTW, even the highest PEAK level, which CAN be detected, cannot be auto detected and "auto set" due to a restriction of the SDK. In other words, if you loaded a sample and found the highest peak level, you would STILL have to MANUALLY set the Target to this level.

As for the Selig Leveler being "near perfect", I agree, with one exception - the ONE thing that does need to be addressed IMO is the visual interface. This is something I'd like to update at some point in the future, to provide more visual feedback in order to aid in identifying the best settings for the job. I feel that this one change could take the Leveler to the next level, even without changing anything else about the device. But first, I have a few other things to tend to…
Selig Audio, LLC

User avatar
Ottostrom
Posts: 845
Joined: 13 May 2016

28 Oct 2017

selig wrote:
28 Oct 2017
If it's working for you, great…but for others…
I disagree, at least in my experience - if you reach down to 16 dB AND use 100% Blend, it probably WILL sound unnatural - but that's not how Leveler was designed to work (certainly with vocals)! I regularly use that much "depth" and get an incredibly natural sound because I only use as much Blend as it takes to get the job done - often 50% or less.

As for noise floor, it's typically 50-60 dB or more below the signal, so reaching down 16 dB should never bring up the noise floor - this is one of the 'features' of the way Leveler works as opposed to other upwards (or downwards for that matter) compressors!

As a general rule, especially with vocals or similar, you'll not get volume consistency unless you set the Curve to include the entire signal dynamic range. So if you set Curve to 3-6 dB, then everything below that will not be affected. If you have a signal that has more than 3-6 dB dynamic range, you will "warp" the dynamics, squishing the top few dB up to the top and leaving the rest down below - effectively creating an EXPANSION rather than COMPRESSION result (but not exactly either one).

So the "key" to being subtle is to use lower BLEND amounts. Think of Curve as simply defining the area you want to affect, and Blend as defining how much affect you want to apply to the output. Blend is VERY much like Ratio in a downwards compressor, such that a setting of 50% Blend is equal to a Ratio of 2:1. Curve, combined with Target, is like the Threshold in a downwards compressor. It's important to remember the results will be TOTALLY different with the Leveler, that is to say the comparisons end there!

So if you want to "level" a performance that covers 15 dB (random number for this example), you MUST set the Curve and Target accordingly. From there you dial in just as much Blend as it takes to bring up the lower parts (when in doubt, use less Blend).
You bring up a lot of good points, and there is a very good chance that I have taken the blend knob for granted in my previous tests. I'm gonna record vocals today anyways so I'll take all of this into consideration when processing it. As always, thanks for the detailed explanations you give Giles!

User avatar
EnochLight
Moderator
Posts: 8405
Joined: 17 Jan 2015
Location: Imladris

28 Oct 2017

selig wrote:
28 Oct 2017
EnochLight wrote:
28 Oct 2017
If I had any feature request for Selig Leveler, it would be to have sample loading for analyzation, so I wouldn't have to play the whole track just for it to detect the lowest level. Not sure how that could be implemented, though. Otherwise, that thing is near perfect.
Once you load a sample, how do you find the lowest level? Technically, the lowest level is noise. The only way to identify the lowest level is to decide where YOU want the lowest level to be. It's not possible to use metering to find this "musical" lowest level, as there are too many variables. It's really a subjective thing.

So unless you know of some existing technology that can detect the softest note/word or a recording, this isn't possible - it's the same probable that prevents us from having a button that does this in real time, so speeding up the process won't make it suddenly possible to do so. Hopefully I'm making sense…

BTW, even the highest PEAK level, which CAN be detected, cannot be auto detected and "auto set" due to a restriction of the SDK. In other words, if you loaded a sample and found the highest peak level, you would STILL have to MANUALLY set the Target to this level.
All great points, hence my "Not sure how that could be implemented, though". Looks like it can't.
selig wrote:
28 Oct 2017
As for the Selig Leveler being "near perfect", I agree, with one exception - the ONE thing that does need to be addressed IMO is the visual interface. This is something I'd like to update at some point in the future, to provide more visual feedback in order to aid in identifying the best settings for the job. I feel that this one change could take the Leveler to the next level, even without changing anything else about the device.
100% agree. Some fancy GUI/UX metering would be choice. Not sure if a waveform display would be adequate, or something else - but it definitely could use some GUI/UX jazzing up.
selig wrote:
28 Oct 2017
But first, I have a few other things to tend to…
#new_device_tease? :D
Win 10 | Ableton Live 11 Suite |  Reason 12 | i7 3770k @ 3.5 Ghz | 16 GB RAM | RME Babyface Pro | Akai MPC Live 2 & Akai Force | Roland System 8, MX1, TB3 | Dreadbox Typhon | Korg Minilogue XD

User avatar
Zac
Posts: 1784
Joined: 19 May 2016
Contact:

28 Oct 2017

I will definitely try that way first from now on Selig. Great explanation.

User avatar
cgijoe
Posts: 77
Joined: 13 Mar 2016

30 Oct 2017

I love this RE so much and use it all the time for sound design and leveling out vocal takes for TV projects I work on.

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

30 Oct 2017

I've got plenty of ideas to get me started, but since I have everyone's attention: any suggestions for how to improve the visual interface? What things would you find useful to aid with setup and understanding what's going on?
:)
Selig Audio, LLC

User avatar
cgijoe
Posts: 77
Joined: 13 Mar 2016

30 Oct 2017

The only thing that ~might~ help out with the visuals, is a graphical display similar to one on Page 6 of the manual. I found that to be particularly illuminating. Maybe even just a sample diagram on the back? I'm not speaking to the amount of work this would take to implement, since I'm no RE developer, just to what I found useful when learning this device. Selig, while I have your attention, just wanted to say you're one of my favorite contributors to the Reason community and your efforts have not gone unnoticed. Just felt like I should throw that out there since its hard to feel positive feedback sometimes. Not enough positivity on these on this rickety ol' world wide web.

User avatar
qn5
Posts: 137
Joined: 29 Nov 2016
Location: NYC
Contact:

31 Oct 2017

selig wrote:
30 Oct 2017
I've got plenty of ideas to get me started, but since I have everyone's attention: any suggestions for how to improve the visual interface? What things would you find useful to aid with setup and understanding what's going on?
:)
Maybe something akin to FabFilters Pro-L where you can see the curves and where it's happening in the waveforms.
QN5 MUSIC
qn5.com

Post Reply
  • Information
  • Who is online

    Users browsing this forum: No registered users and 20 guests