Jump to content

Normalize clip to RMS/LUFS


Jyri T.

Recommended Posts

This is an edited/updated version of the original post. The following 4 posts refer to the original one.

Hi bakers!

I would really like to see a feature to normalize clips to a certain RMS/LUFS-level in an easy and fast way so that you don't have to do it separately (read: tediously) with a third-party plug-in.

It should be simple enough as it only needs a few parameters.

  • Normalize to: X
  • Normalize by: peak/RMS/LU/K-system/whatever
  • If overs: Reduce gain/clip/limit/cry for help

There could be some advanced options as well.

  • Normalize to (RMS only): Average/peak
  • RMS time window (RMS only): (default: 700ms)
  • Merge clips on same track? Keep separate/Merge together
  • Apply clip trimming before normalizing? (default: Yes)
  • Do I love Cakewalk? Yes/Indeed

This way you could easily set up a project with a schnizzleton of clips with audio levels all over the place to a relatively smooth starting point in no time.

Now you need to do this manually and I still don't understand why you have to do it in Sonar since it really is no difficult feat to program and I'm sure many of us would appreciate this.

PS. If normalizing is NOT your thing, feel free to keep it to yourself. This thread is NOT called "Normalizing – good or bad".

Edited by Jyri T.
  • Like 3
Link to comment
Share on other sites

3 hours ago, Starship Krupa said:

If you select the clip you wish to normalize, then from the Global menu, select Process/Apply Effect/Normalize, does that not do what you want it to?

(this action is also available as a keystroke binding)

Yes I was about to say the same thing...select clips

Press Alt - Shift - N or use the menu

Normalize.gif

 

image.png

Edited by Mark MoreThan-Shaw
Link to comment
Share on other sites

One thing to bear in mind when using Process->Normalize  / ALT + SHIFT + N, is that you should consider apply trimming or bounce to clip(s) to each clip individually before normalizing it. 

The reason being, is that when a clip is split or slip-edited, the remaining parts of the clip are still there even though hidden, and will be included in what is being considered when normalizing. This is especially true for split clip, where you've actually got two copies of the entire clip when it's split.

Of course, after apply trimming or bounce to clip(s), you won't be able to slip edit the hidden parts back in any more.

  • Like 1
  • Thanks 2
Link to comment
Share on other sites

Oh zut alors!

I didn't recall the Normalize function DOES keep the clips separate. Sorry! My bad!

Yet the main point is the ability to normalize by RMS or LUFS. The inbaked function ONLY has the peak value option and it's not enough any more IMH-ishO.

Jyri

Edited by Jyri T.
Link to comment
Share on other sites

  • 2 months later...

I surprised nobody point this out. 
 

Peak level and LUFS/RMS are completely different things. 
 

Peak level can be controlled with simple attenuation or volume/ level controls. This is none destructive. 

Normalize takes the level to a pre set peak value and destructively renders the audio. 

But one tiny transient in a clip or file means the whole thing is only normalized to that one peak. 

LUFS / RMS are effected by peak level up to that max peak. You can turn up the audio until it’s maxed out at   -.01 db and that’s as loud as it can go. Beyond that you have to apply compression/limiting. Which is none destructive. 

What you’re asking for is a Compressor/Limiters that is destructive. ?

Mastering is an art form. It is one of the highest regarded fields in the audio industry. It’s getting easier to DIY but for best results I still highly recommend studying the topic and understanding your tools that have now been handed to the masses free. 😮
 

Edited by John Vere
  • Like 3
Link to comment
Share on other sites

  • 1 year later...

I have been looking for a plugin that easily does the following: Normalize a clip (or, better still, a bunch of individual clips) to, say, -20 LUFS. Easily meaning it uses some direct access like ARA (i.e. not needing any "listening" of the audio) and thus being easily used on a large group of clips.

No luck.

It should NOT be a big thing to code. So, the question remains, WHY IS IT NOT YET AVAILABLE IN CAKEWALK?

I really don't understand why you can't bake it.

Slava Ukraini!

Link to comment
Share on other sites

select clip(s), select "process -> apply effects -> normalize" (or gain depending) and set the parameters and hit ok.

image.png.1e2a2cb4ed4f3f71866679f6bd986c4f.png

each clip selected will have the normalization performed.

depending on your goal, clip gain can be very useful for leveling and overall changing level as well

image.png.ccc142228231be3ed6b5f80d3634a219.png

Edited by fossile
Link to comment
Share on other sites

I think one of the issues is that LUFs and other loudness algorithms have a time associated with them. You usually don't just say, I want this whole song to be at -20LUFS.  You generally want the loudest part of the song to be at some loudness limit. The loudness will vary as the song goes along. Therefore, the program would have to analyze the whole song file to find the loudest part and then figure out what gain, compression, limiting etc. to do to get the loudest part at the desired loudness.

Izotope's Ozone actually does some of this but as you play with it you see that it is more dynamic then just saying "Make this whole song this level".

Link to comment
Share on other sites

definitely - there are really three levels measured in the LUFS/LU - the "integrated" (you need to be at or below the spec at the end of the total sound recording).  "short term" (which has a time window 3000ms), and "momentary" (very short measurement window) 🙂 then you have the loudness range (usually in LU) which gives you some idea of the dynamic range. 

one thing i've observed in mastering forums and online tutorials, blogs etc is that many mastering engineers (i am not one, nor do i play one anywhere) will do a bunch of their magic to get the integrated value to be where it needs to be (mainly critical for broadcast because in some countries there can be monetary penalties) and to accommodate the behaviors of streaming services automatic leveling "feature", and then leverage higher short term and momentary levels to get the overall level to be perceived as "loud" without violating the integrated requirement. 

definitely different than simply looking at peaks and RMS. the Izotope Insights meter is very nice, the TC metering is also very nice, and the free YouLEAN meter is decent. some free ones i've played with vary in their accuracy.

Edited by fossile
Link to comment
Share on other sites

21 hours ago, reginaldStjohn said:

I think one of the issues is that LUFs and other loudness algorithms have a time associated with them. You usually don't just say, I want this whole song to be at -20LUFS.  You generally want the loudest part of the song to be at some loudness limit. The loudness will vary as the song goes along. Therefore, the program would have to analyze the whole song file to find the loudest part and then figure out what gain, compression, limiting etc. to do to get the loudest part at the desired loudness.

Izotope's Ozone actually does some of this but as you play with it you see that it is more dynamic then just saying "Make this whole song this level".

This is NOT what I'm talking about.

I'm talking about starting a project with a ton of clips that are all over the place when it comes to audio levels.

I want to be able to get all the individual clips easily on a consistent level at the very beginning, before I start editing them and mixing the song (or what ever project I'm working on).

That's what I'm talking about. Individual clips, not track or bus output.

Link to comment
Share on other sites

 

5 hours ago, Jyri T. said:

I want to be able to get all the individual clips easily on a consistent level at the very beginning, before I start editing them and mixing the song (or what ever project I'm working on).

yes, so either prep them using a tool to adjust gain to a consistent level, then import. or in the DAW, use clip gain to adjust the entire clip or just parts of it. or if destructive gain setting is ok, then use the process->apply effects->gain and set it up or down. 

Link to comment
Share on other sites

19 hours ago, fossile said:

 

yes, so either prep them using a tool to adjust gain to a consistent level, then import. or in the DAW, use clip gain to adjust the entire clip or just parts of it. or if destructive gain setting is ok, then use the process->apply effects->gain and set it up or down. 

If I prep them beforehands, it applies to the whole files. I may use only parts of them so the prepping may not help.

And applying gain manually on individual clips is exactly what I want to avoid here.

  • Like 1
Link to comment
Share on other sites

unfortunately i don't think here is a generally automated way to get reliable results. one thing i use is the Melodyne level balance macro to increase quiet sounds and reduce louder sounds: https://helpcenter.celemony.com/M5/doc/melodyneStudio5/en/M5tour_MacroLeveling?env=standAlone

i believe it's available in all versions except the essentials.

Link to comment
Share on other sites

On 1/20/2023 at 6:21 PM, Jyri T. said:

This is NOT what I'm talking about.

I'm talking about starting a project with a ton of clips that are all over the place when it comes to audio levels.

I want to be able to get all the individual clips easily on a consistent level at the very beginning, before I start editing them and mixing the song (or what ever project I'm working on).

That's what I'm talking about. Individual clips, not track or bus output.

This^, I like it!

I had the same issue. Especially when you do comping it is very important not to be misled by the clips volume difference. Thus before I start comping, I level all clips to the same LUFS value. Also, when you recorded some parts of the song separately, it is helpful if all of them match the same LUFS value. It makes mixing much easier!

I always wondered how people could use the CbB normalize function for this (peak based), because if you have one single spike in the audio, then this clip has a totally other volume, than others!

Today I do it in CbB like this:

  1. I use the free HOFA 4U Meter Fader to find out the LUFS value of the clip (within CbB just drag the clip onto the plugin).
  2. Then I use "Process > Apply Effect > Gain..." to adjust the value to my desire.

This is not very elegant, but it works. Surely it would be better if there was such a function in CbB, I agree! One thing I want to mention to the HOFA 4U Meter is that if your file is mono, then it shows the LUFS value as double mono (+3dB). But this is easy to consider.

Recently I also use Reaper to achieve this. In Reaper there is a "Batch File/Item Converter" that can apply a LUFS volume adjustment and/or any plugins to a batch of files. Very helpful!

  • Like 2
Link to comment
Share on other sites

4 hours ago, Light Grenade said:

Check out The Normalizer by HorNET, it can do auto-normalizing with the grouping function.

Low-cost and fast. I use it all the time. 

I tried this one, too! Sometimes it generates very strange spikes in the beginning of the clips and on certain ess sounds! Unusable IMO!

Link to comment
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...