There are too many variables to give a reliable answer but assume that whatever visual or audio work you put into a project, less than a few percent of people, if you are lucky, will see or hear the results as you will do. That is assuming everyone's eyesight and hearing is roughly equal. Then take into account their viewing/listening environment. Then the sophistication of the devices used for playback. Then how those devices have been set up, and by that I mean the user adjustable controls in the pieces of equipment setup menus. Then lastly the volume levels that are chosen at the time of viewing. Even if using headphones there is room for variation dependant on the type of headphone in use and surrounding noise environment.

[Edit].

Then look at it backwards. What is your hearing, visual acuity and colour vision like to make initial adjustments to a project? Add all of that together and the answer is:- It's a lottery.

How about if you listened using the same volume level as you used when editing/mixing it? Would or could that give an accurate result?

Hi,

simply use the mixeur (type M on the keyboard) to see each sound track in your video and the result output level. Normally the output must be around -6 decibels to obtain a good result. Also you can adjust each track from sound curve to the right level.

Regards,

YR

@Mitch-way06

Hi

As @CubeAce has said there are too many variables and @yvon-robert is correct in setting the peak level to -6 dB, however there some considerations/steps to take into account/do - I works as follows:

1. Normalise all audio for both video and background sound effects etc
    
2. Set the volume of the Background effects to somewhere between -12 and -20 dB - right click Audio functions, Set volume
    
3. Set the volume of the Foreground audio eg people talking other close sounds set to the volume level, as in 2, so that sound level is correct relative to the background and the closeness of the sound source/type etc.
    
4. In the Mixer, adjust the Master volume controls - see note below - so that the loudest audio peaks at -6dB while playing.
    
5. Where necessary, use volume curves to alter the volume of audio eg music to introduce ducking in the levels where you want for example speech to be more clear.

Note: I and several other users, use the Melda Production MDynamiqEq for limiting - it find it gives a better clarity and is more versatile than the MEP effects and you do not need to use the Master volume controls to set the maximum level.

The rest is up to the viewer and their TV/player volume control.

HTH

John EB

@Mitch-way06

I can only wholeheartedly endorse everything that @CubeAce and @johnebaker have written.

In my case I tend to use iZotope Ozone 8 Advanced plug-in as my "equivalent" to John EB's use of the Melda plugin he mentions.

Jeff

How about if you listened using the same volume level as you used when editing/mixing it? Would or could that give an accurate result?

@Mitch-way06

It would work for you but maybe not for me. As I get older, my hearing across the frequency bands is changing. Particularly, my right ear is somewhat less responsive than my left ear around the upper frequencies, so anything say placed centrally in your mix may appear slightly to the right when I listen to it.

I may not even be able to hear all the frequencies as clearly as someone much younger.

This is also important to know when mixing down. I often look at levels of individual channels and relative pan positions. If I'm not sure, I'll pan the sound to my left ear for verification of volume before panning it to it's respective postion to the right and prey the developers have got their settings correct. The ability to solo a channel for this is also nice to have.

Our hearing is never equal between left and right. Sound is as individual as sight is to any one person. And as with sight, most of the processing happens in the brain rather than at the organ responsible for the input. People can train themselves to see better and to hear better if they apply themselves, but that in itself can ruin any enjoyment of the experience. When that happens, you have possibly turned a hobby into work. 😉

One method might be that before saving or exporting, check all of your audio tracks on the timeline, play from there one by one to make sure you can hear it. Then adjust the volume of each track by right click on the audio clip, click set volume, and adjust to desired level. If you can hear it, they can too, if not, neither can they. I do this sometimes on videos with only a few audio clips or one. Also I think I saw a "normalize" option in settings somewhere for auto adjustment of all audio volumes to be level or equal. If only one track audio with different volume levels of sound it keeps the volume level consistent throughout the entire clip/track.

@russelldean

Welcome to the Magix forums.

 I think I saw a "normalize" option in settings somewhere for auto adjustment of all audio volumes to be level or equal. If only one track audio with different volume levels of sound it keeps the volume level consistent throughout the entire clip/track.

I apologise in advance, but your description of what Normalise does is factually incorrect, though it is a common belief!

Normalise is the process whereby the peak (loudest) level of any audio object is set to whatever the user wishes, often 0dB or -1dB or similar. In doing this no change is made to the relative levels of any audio within that object. In other words, if there is, say, a 20dB difference between the "loudest" and quietest sounds on a track and the peak before normalisation is -6db, "normalising", e.g. raising the peak to -1dB will still mean that difference between loudest and quietest will be 20dB. Normalising does not and cannot alter in any way the dynamic range of the audio file; only the use of compression or limiting can achieve that alteration.

One other point: if you apply the Normalise process to a complete audio track, perhaps containing several audio objects, each object will be normalised to the level the user chooses. So, a "loud" clip may actually change very little, indeed it could become quieter if its current peak level is higher than the normalisation setting; a "quiet" clip, on the other hand, will become noticeably louder. These changes to the individual objects may be desirable, or they may not!

Jeff
Forum Moderator

@russelldean

Welcome to the Magix forums.

 I think I saw a "normalize" option in settings somewhere for auto adjustment of all audio volumes to be level or equal. If only one track audio with different volume levels of sound it keeps the volume level consistent throughout the entire clip/track.

I apologise in advance, but your description of what Normalise does is factually incorrect, though it is a common belief!

Normalise is the process whereby the peak (loudest) level of any audio object is set to whatever the user wishes, often 0dB or -1dB or similar. In doing this no change is made to the relative levels of any audio within that object. In other words, if there is, say, a 20dB difference between the "loudest" and quietest sounds on a track and the peak before normalisation is -6db, "normalising", e.g. raising the peak to -1dB will still mean that difference between loudest and quietest will be 20dB. Normalising does not and cannot alter in any way the dynamic range of the audio file; only the use of compression or limiting can achieve that alteration.

One other point: if you apply the Normalise process to a complete audio track, perhaps containing several audio objects, each object will be normalised to the level the user chooses. So, a "loud" clip may actually change very little, indeed it could become quieter if its current peak level is higher than the normalisation setting; a "quiet" clip, on the other hand, will become noticeably louder. These changes to the individual objects may be desirable, or they may not!

Jeff
Forum Moderator

So how does this affect listening with a low volume? Does normalising mean that you will still be able to hear the quiet sounds even if you’re watching with a low TV or headphones volume?

@Mitch-way06

No, it can be quite the opposite. emmrecs (Jeff) has the correct explanation of normalizing.

@Mitch-way06

Hi

. . . . Does normalising mean that you will still be able to hear the quiet sounds . . . .

To clarify with two examples all dB figures quoted (unless otherwise stated ) are relative to 0dB. Normalisation is also set to 0dB.

Case 1.

Audio peak is at -6dB, quietest is -20dB before normalisation - normalisation will change the levels to 0dB and -14dB respectively - ie the audio is now louder by +6dB for the peak and the quietest part, their relative levels to each other are still the same.

Case 2

Audio peak is at +6dB, quietest is -20dB before normalisation - normalisation will change the levels to 0dB and -26dB respectively - ie the audio is now quieter by -6dB for the peak and quietest part.

However if the difference between the loudest peak and the quietest part is much larger ie the quite is at -30dB then it is possible that the quiet part is reduced by an amount such that it is now at a level which is close to or below the audio threshold for a given volume setting on play back.

As an approximate guide a 10 dB difference between level is a doubling or halving in loudness.

HTH

John EB

@Mitch-way06

You want to use compression.

See this thread where we discussed it recently.

This applies where both the loud and soft audio are in the same object.  From the manual, with compression, "loud passages become quieter and the total value is raised. This makes the volume more consistent and speech easier to understand. A compressor is mainly useful in case background noise or music interferes with speech and simply increasing the volume of the individual objects or tracks does not result in any significant improvement."

Normalize and compress, then maybe normalize again or raise the gain to compensate because compression reduces the gain. This reduces the louder parts so that they are closer to the softer parts, above a specified threshold. A graph that illustrates this would be helpful. Tomorrow.

There is a difference between having the louder parts with one video clip, and the softer with another. In that case you could normalize both and then reduce the clip with the softer audio to sound like it's more in the background.

Each case is different.

John CB

@Mitch-way06

You want to use compression.

See this thread where we discussed it recently.

This applies where both the loud and soft audio are in the same object.  From the manual, with compression, "loud passages become quieter and the total value is raised. This makes the volume more consistent and speech easier to understand. A compressor is mainly useful in case background noise or music interferes with speech and simply increasing the volume of the individual objects or tracks does not result in any significant improvement."

Normalize and compress, then maybe normalize again or raise the gain to compensate because compression reduces the gain. This reduces the louder parts so that they are closer to the softer parts, above a specified threshold. A graph that illustrates this would be helpful. Tomorrow.

There is a difference between having the louder parts with one video clip, and the softer with another. In that case you could normalize both and then reduce the clip with the softer audio to sound like it's more in the background.

Each case is different.

John CB

Hi,

Yeah well in my case it’s the quiet sound effects in the distance that need to probably be louder, not the dialogue. It’s just if you happen to be watching with a lower volume, those quiet sounds cannot be heard. That’s why I was wondering whether there is something you’re meant to do to make it so that the quiet sounds can still be heard when listening with a lower volume. Like if it’s late at night and other people in the house are asleep, you wouldn’t want to disturb them so you have to watch TV with a lower volume don’t you.

@Mitch-way06

Since we can't hear your audio, we don't know what your timeline looks like, so we have to guess. There is no set answer, which is what I think you want.

Try this. Make sure that your audio is on a separate track from your video. If not do so (Ctrl+H). Open the Mixer. If your audio is on track 2, click on the FX button in the Mixer for track 2.

Turn on the compressor and try the settings shown below.

Make sure to increase the gain to get the loudest part back to about the same level. The quieter parts will now be louder. After that, it's a matter of playing with the threshold and compression, with make up gain to get to the same max. level, not louder.

You can turn the compressor off and on to listen to before and after.

John CB

@Mitch-way06

Since we can't hear your audio, we don't know what your timeline looks like, so we have to guess. There is no set answer, which is what I think you want.

Try this. Make sure that your audio is on a separate track from your video. If not do so (Ctrl+H). Open the Mixer. If your audio is on track 2, click on the FX button in the Mixer for track 2.

Turn on the compressor and try the settings shown below.

Make sure to increase the gain to get the loudest part back to about the same level. The quieter parts will now be louder. After that, it's a matter of playing with the threshold and compression, with make up gain to get to the same max. level, not louder.

You can turn the compressor off and on to listen to before and after.

John CB

When you say “try the settings shown below”, do you mean use those exact values as shown in your screenshot?

Yes

@Mitch-way06

Results? Did you find any settings to help? Did you play back and listen with the volume turned down to see if the result was closer to what you want?

The images below show the curves for compression - impact of threshold and ratio, along with makeup gain to get back close to 0. The illustration is for threshold at -20dB, ratio 2:1, not the values that I told you to use.

John CB

@Mitch-way06

Hi

. . . . I was wondering whether there is something you’re meant to do to make it so that the quiet sounds can still be heard when listening with a lower volume. Like if it’s late at night and other people in the house are asleep, you wouldn’t want to disturb them so you have to watch TV with a lower volume don’t you . . . .

IMHO this is a Catch 22 situation - if you adjust the relative levels so you can hear the quiet audio with the volume of the TV turned down then when listening at 'normal level' the background sound will be too loud. Once the video is exported or a disc burned the ratio of quiet to loud is fixed.

The playback device has to do the changing of the relative levels and it would need an adjustable compander to do this.

I use Bluetooth headphones on my TV for listening when people are asleep, no issue with relative levels.

HTH

John EB

@Mitch-way06

Hi

. . . . I was wondering whether there is something you’re meant to do to make it so that the quiet sounds can still be heard when listening with a lower volume. Like if it’s late at night and other people in the house are asleep, you wouldn’t want to disturb them so you have to watch TV with a lower volume don’t you . . . .

IMHO this is a Catch 22 situation - if you adjust the relative levels so you can hear the quiet audio with the volume of the TV turned down then when listening at 'normal level' the background sound will be too loud. Once the video is exported or a disc burned the ratio of quiet to loud is fixed.

The playback device has to do the changing of the relative levels and it would need an adjustable compander to do this.

I use Bluetooth headphones on my TV for listening when people are asleep, no issue with relative levels.

HTH

John EB

 

 

How about if you made two different versions of the same video? As in make one version with the audio optimized for listening with a lower volume, and another version with the audio optimized for listening with a higher volume? Could that be a solution?

@Mitch-way06

Two versions would do it, but first, try to get one that has the background sound closer to the foreground, but not too much. You may find that it improves the audio at all volumes. If not, you still have to find a solution for the quieter version. How are you doing with this?

John CB

@Mitch-way06

Two versions would do it, but first, try to get one that has the background sound closer to the foreground, but not too much. You may find that it improves the audio at all volumes. If not, you still have to find a solution for the quieter version. How are you doing with this?

John CB

Hello John,

How would you suggest I go about making the background sound closer to the foreground, as you say?

Also, I haven’t yet tried out the settings that you showed me for the compressor. I will let you know the result of that once I’ve done it.

@Mitch-way06

How would you suggest I go about making the background sound closer to the foreground, as you say?

 I haven’t yet tried out the settings that you showed me for the compressor.

When you do try out those settings in that second sentence, you will almost certainly have the answer to the question in your first one!

Jeff

@Mitch-way06

Hi

. . . . How would you suggest I go about making the background sound closer to the foreground, as you say? . . . .

1. If the background sounds are separate audio files from the foreground audio then try the method I described above.
    
2. If all the audio contains both background and foreground sounds then you are going to have to use the compressor method as described by @browj2 - you are going to have to experiment with the compressor settings on those files.
    
3. If there is a mix of items 1 and 2 then both methods may need to be used.

Do be aware there is no quick and easy one step fix, and it can be time consuming getting the balance right for the normal and quiet versions

HTH

John EB