Normalizing audio simply refers to the process of applying a constant amount of gain to audio in order to increase the amplitude to a target level.
In simpler terms, when you normalize an audio track, the samples in the file move up by the same amount, so that the highest sample value becomes 0 db or close to 0 db. Which means, if your sample is already near that, you won’t notice any increase in volume.
So, Should you normalize audio before mixing?
Normalizing should not be carried out before mixing because it can be destructive to your mix and the goal of mixing is not to increase the volume of your audio track peaks but to simply blend audio tracks with other tracks in the session. Normalizing audio is more applicable in mastering because it can help you get your audio all the way up to your digital ceiling without using compression or a limiting. Even then, normalization doesn’t offer enough flexibility for you to rely on it solely.
Other people may consider normalizing before mixing but I choose to avoid it because it may cause several issues.
It is more applicable to mastering because normalizing can help you get a good boost in volume uniformly but it has to be used in moderation because audio is pretty delicate when it comes to it’s handling.
When to use Normalization
In the modern standard, seeing as we have so much development in digital audio Workstation technology.
Gain staging is a much better option nowadays because it less complicated and can help you out in your mixing to get back some volume.
One of the best applications of normalization that I personally utilize is with samples. Working with samples often times means working with a low quality song due to the fact that it was recorded, mixed and mastered using old technology and techniques. To be able to get some uniform volume and brighten up the sample, normalization works very effectively.
The good part is, you’ll find the normalization feature in all modern DAW’s, so you can basically read this article and try to work with a sample using what I’ve prescribed here.
On most modern samplers, you’ll find the normalize function because sampling really benefits a lot from normalization.
I also use normalization on Kicks and any other one-shot samples like snares, rim shots etc. that may seem too shallow, purely giving an indication that normalization can work to give them a good boost.
The advantage with this is that I can easily keep an eye on a single kick drum or snare in the mix. I can easily control peaks and use a soft clipper to ensure that I don’t end up with a distorting or clipping instrument.
For situations like this normalization is quite applicable because you’re basically working with a one-shot and the dynamics are pretty much manageable and don’t span over a long sample.
Why normalization may be bad for your mix
Loss of originality
Normalization can work best and help you get a good boost but as with anything, it can also be bad for a mix.
If you normalize an audio track and save it while completely discarding the un-normalized original version you risk completely losing the original which could be bad if you realise that normalization might not have been the best way to go about things.
There are a lot of practices that can be destructive in digital audio Workstations and one of them is the inability to undo certain things once you save changes.
This is one of the reasons why I don’t personally recommend using normalization before mixing especially to newbies that are still figuring out their way around mixing; Because it is pretty sad to see somebody make a change to track that they can’t undo.
Distortion
It’s no secret that a boost in volume also bears the risk of eventually running up your audio into the realm of distortion.
With normalization, you need to understand that once it is applied, the highest peaks in your audio will be raised to the target level which is 0 db or close to it.
When the peaks are raised things get pretty much unpredictable. I say this because when your digital audio gets converted to analog in order for it to play through your speakers, the filters pretty much reconstruct the signal. This can pretty much lead to inter sample peaks which would result in your audio clipping leading to distortion and harshness.
Gain staging
Now that you have a grasp of normalization and how it works to boost audio. we can now have a look at what I prefer to use when I need to give some volume to an audio track.
It’s called Gain Staging.
All gain-staging essentially does is that it properly control the levels of your audio signal flow inside your DAW to avoid clipping and distortion.
You do this by checking the volume of each element you record to make sure that it doesn’t exceed the level you consider healthy for your mix.
Gainstaging is primarily important on your master bus and I usually set the tracks peaks to around -9db and the body of the waveform at around -18db.
Having enough headroom for your audio tracks will give you the choice to increase and boost your gain at the master fader without clipping.
Should I use Normalization in Mastering?
The problem with going the normalization route in mastering is that you’ll not have enough flexibility to work with.
When it comes to mastering, we want to get the appropriate amount of level without distortion.
Furthermore, we want to get this level to sound as natural as possible.
This is where most people will usually go for a limiter or compressor because they can give you enough level and also give you subtle adjustments to the dynamics which could improve the overall final product.
Compression and limiting require good knowledge of both to effectively employ them in your mastering, and you’re better off learning how to use tools like them to master your audio rather than rely on normalization as a one hit trick for a boost in volume.
