@Tony-Haack

Hi

Not John CB, however I too use the Youlean loudness Meter and your levels in the above images look good to me 👍

John EB

@CubeAce

Hi Ray,

This is somewhat off-topic.

Regardless of using a luffs meter or not I have found that the closer you get your recordings to peak to within .05db the less YouTube will re-encode and compress the sound of the finished video. The more YouTube does that, the more variation there is in volume between each video uploaded. Ideally I have found I need the 'Stats for nerds' to read 100/ 100 for volume consistency between videos. Very few uploads achieve that but seem to when the largest peak within the file is -.05db. 

Please read this article.

YT uses both LUFS and maximum loudness. If the video exceeds -14 LUFS, YT will normalize it, meaning that it will reduce the volume until the video measures -14 LUFS. If you have clipping, YT will also reduce the volume. You should not have clipping in any video or audio recording.

Where do you get this "0.05db" from? You can't even measure to hundredths of a dB; 10ths only.

In Stats for Nerds, you will get 100/100 if your audio is less than -14 LUFS when uploaded - nothing to do with the maximum instantaneous sound. Check Content Loudness for this. Also you do not know by how much you were were below -14 LUFS.

In the image below, I made sure that the audio in this video did not exceed -14 LUFS. This is what shows in YT:

100/100% (content loudness -2.1dB.

Here is the output from YouLean:

Actual was -16 LUFS with True Peak Max of -0.7dB. I could have increased the output to get closer to -14 LUFS but then I would have had a lot of clipping as you can see by the red indicators at the top of the chart.

This was with music. With just narration, you are not going to easily get to -14 LUFS without a lot of work, removing peaks and compression.

Without measuring LUFS, you don't know what you have.

@Tony-Haack

Hi Tony,

Looking at your graph, you have a True Peak Max of -5.8 dB, which means that you can easily raise the level of your audio by 4.8 dB without going over the -1 dB recommended maximum or clipping.

However, if your target is actually -23 LUFS, then you should leave the audio as is.

If your target is louder, say closer to -14 LUFS for posting on YouTube, then you can easily raise the overall level.

I presume that you have more than one recording. What happens from one to the other? How does it sound to you?

Do you have other audio, such as audio with the video or music? Are they louder than your narration?

John CB

@Tony-Haack

Further to my previous post, there are a lot of factors to consider about audio and getting the narration right is just one of them.

The main objectives are:

1. Getting a consistent and appropriate perceived loudness for the overall audio
2. Getting consistent loudness from one audio recording to another
3. Getting consistency from one project to another.

What one wants to avoid is having to turn the volume up and down within the same video because some parts were too loud, others too soft, and having to adjust the volume from one video project to another. When you listen to your TV or YouTube, you usually set a comfortable level and maintain that from one show to another. Advertisers have always tried to get their audio to be louder than the program and that has now been made less intrusive because of the standards for perceived loudness.

Here is a test that you can do.

Start a test project, add some video clips that have sound, add or create some narration but for several seconds each at different locations with gaps between them, then add some music.

There are some free songs that come with MMS under the Audio tab. Import Sunrise and then Reunite onto track 5, starting at the beginning of the project. Trim each down to about 15-20 seconds each. Look at the waveform of each song. They are quite different.

So you now have video sound, narration and music all playing at the same time.

Put YouLean on track 5. Select ASWG-R001 Portable instead of Default. This will set the target to -18 LUFS.

Playback and watch what happens. Sunrise will come in at about -19 LUFS, but when you hit Reunite, it goes way up to -9.7 LUFS and what you hear sounds much louder than the first song. Just playback Reunite and it will show about -8.4 LUFS. You want a fairly constant perceived loudness from one song to the other so you would reduce the volume of Reunite by a lot and slightly increase Sunrise.

Put YouLean on your narration. It's most likely way below -10 LUFS.

Now, if you want to hear the narration, you'll have to duck the audio. With the music at the import levels, you would reduce the volume of Sunrise by 9-12 dB to properly hear your narration, but you would have to reduce Reunite by at least 22 dB to properly hear your narration.

Check the Reunite part with YouLean on the Master FX strip, and the combined integrated loudness will climb a bit more. Sounds terrible.

Use maybe 5 seconds of narration and put it in the middle of the first song, copy that to the middle of the second song. Right click on the first narration, decrease volume and select the default of 9 dB, ok. You'll get a volume reduction of all other audio - Sunrise and your original audio with the video. Right click the second narration under Reunite, Decrease volume but select 12 dB. Now play back all with YouLean on the Master. The first one probably sounds fine to you but the second one will still have the music playing too loud or louder than the one with Sunrise. So, Reunite definitely needs its volume reduced to get consistency. Then, when I do the ducking, the 12 dB drop should work.

The above image demonstrates what I mean.

BTW, I normalized the audio of the narration and then reduced the gain by a little more than 1 dB.

This should give you more ideas as to how to make adjustments to get the sound more consistent.

We haven't discussed the original audio of the video clip. This, too, needs to be addressed. Often, I just mute the parts not needed. Still, you may have to modify the original volume and you now have the tools to check what is happening.

Lastly, since you started this thread with ME3, your original audio may need some cleaning up, and that could be where you would want to use the external audio editor. Unfortunately, ME3 does not have Spectral Cleaning which is great for removing certain sounds. Spectral Cleaning is in SF Audio Cleaning Lab. However, the cleaning tools that are in ME3 are somewhat more sophisticated than those in Movie Studio under Audio Cleaning, and you can use VST plugin effects.

By playing around like this, especially with some extremes, you'll get a better understanding as how to make adjustments. Use your ears and eyes (on the instruments and waveforms).

John CB

Thank you, John - the tasks you have set me are in progress - not fully nailed, but underway. Earlier you mentioned "Ducking". Is this feature in Magix Studio 2023? If so, where?

@Tony-Haack

Hi Tony.

I see John CB has not replied to this yet. To get some-ones attention you need to select them by pressing @ on your keyboard and selecting them from the list that appears so you get the @browj2 which then sends them an email.

To do 'Auto Ducking' you need a side chain compressor which MMS or VPX does not have. You could try downloading a free one from here. First make sure you read the limitations of each one to see which you need. I personally haven't used a side chain compressor since they were first available in hardware in the 70s and used via a real mixing desk to analogue tape.

You need to add it to the vocal track which then becomes the master track to 'duck' the track you wish to lower the volume on. The track you wish to 'duck' has to then be connected to that track and the volume set to the desired reduction via the controls on the Side Chain Compressor.

Ray.

@Tony-Haack, @CubeAce

Hi

. . . . To do 'Auto Ducking' you need a side chain compressor which MMS or VPX does not have. . . .

The Auto ducking @browj2 mentioned, is in the recording options for the Audio recording option in MMS under the advanced button with options for setting the amount fade down/up and if it is to affect the video or all audio objects which are located at the same point in time in the project.

John EB

@John EB

John - I had missed the advance tab option for Ducking - had been doing each section manually. Thank you.

@Tony-Haack

Hi Tony,

Watch this:

Unless you are very good at doing narration and the location of the recording when you do it is perfect and nothing will be moved afterwards, do not use the Ducking feature shown by John EB under the Advanced button of the Recording dialogue.

As I mention in the video, do ducking at the very end when nothing will be moved, otherwise, you will have to fix the volume curves where the sound was reduced that would now be at the wrong location. Understanding Work Flow is important.

If you have 2 or more audio narrations that are placed together because of editing the way that you recorded - and not just one object, just duck them all. Try it. It will work fine.

John CB

@Tony-Haack, @browj2

Hi

Perhaps I did not make it clear that I was responding to Rays comment I quoted.

As I said in my other post here, I do not use the auto ducking, mainly because my workflow is somewhat different.

Hope this clarifies my comment above.

John EB

@johnebaker

Hi John EB,

Perhaps I did not make it clear that I was responding to Rays comment I quoted.

Yes, I realized that, just wanted to point out that's it's not usually a good idea to do automatic ducking, especially in Tony's case after seeing his descriptions of recording and editing process, similar to mine. He would likely end up with a mess; I know that I would.

@Tony-Haack

Hi Tony,

See also your other thread about the same/similar topic.

John CB

@johnebaker 

@browj2

@CubeAce

I have just completed testing again -

1) Edited an audio object on the timeline

2) Split it into various Parts

3) HIghlighted those parts and exported then as a WA file

4) Dragged the exported file back on to the timeline

5) It was all joined together.

6) Bottom line - didn't need to use Music Editor 3 - solution - simple.

May thanks for all of your inputs - now have a workable solution to move forward - learned a lot on the way.

Tony

I'm trying to externally edit an audio clip but I'm getting an error message with Music Editor 3. Error code -6. Possible that the installation folder is damaged. I'm unable to find how to reinstall the program. Do I need to reinstall Movie Studio 2023?

@dgclayton

Hi

Does the Music Editor start if you browse to the folder:

C:\Program Files\MAGIX\Movie Studio 2023 edition \MusicEditor

and double click on the file MusicEditor.exe

If not try disabling your Antivirus and test again. If the program starts then you need to add an exception for Music Editor in you Antivirus Realtime/Folder/Ransomware feature - how you do this will depend and the Antivirus program.

Otherwise need more information to help diagnose the issue(s) you are having:

The specification of your computer system including Windows version and program version, see this topic for what is required and please quote processor and graphics card make/model in full, and also what monitor/screen resolution(s) you are working with if this is a laptop.

I would suggest you put this information in your profile signature so we do not have to keep asking for it.
 
Did you get any error messages when installing MMS 2023 and/or are getting any other error messages, if so then please quote them in full or provide screen shots.
 
Are the graphics card/integrated GPU drivers up to date?

John EB
Forum Moderator