Let's say you've just completed a perfect mix full of your latest and greatest tunes and you want to use it as a promo tool. If you plan to send your mix out, it really needs to be at its best.
With this in mind, let's take a look at some things you do to ensure your mix is firing on all cylinders. We'll cover manual editing, some basic mastering methods and essential conversion techniques.
Once you've recorded your mix, you should be left with a file. It might be WAV, it maybe AIFF, this is not hugely important. What is important however is that it's in a resolution and bit depth of at least 16 bit / 44.1 kHz and is not distorted beyond the point of recognition.
So whether you are recording your mix live or in the studio, try to ensure you have plenty of headroom and are hitting the disk at a decent resolution. I would suggest 24 bit / 44.1 kHz as this will give us the most flexibility later.
In this case, the recording was done correctly but you can see that the level is actually a little low. You can either raise the level by a specific amount or normalize it. Luckily, there was no obvious noise in the recording so this wasn't amplified with the boost.
Our original DJ Mix recording at a very low level.
Once boosted, you should start to get a decent overview of the file. Make sure there are no obvious faults, or dropouts at this point. It won't hurt to give the whole thing a listen. Once you are happy... You can move on and start to think about some processing.
The same mix with straightforward normalization applied.
The first thing to sort here is the overall level of the mix. This really is key to be honest. It's quite likely that there are some pretty serious peaks and troughs, especially where you have mixed a couple of tracks or added effects on the fly.
We will be using some dynamics processing later but for reasons that will become clear, this simply can't be overdone. Before we apply any plug-ins, I like to manually reduce the really obvious peaks. This can be done using automation in your DAW or by careful selection in a dedicated audio editor.
Manually reducing peaks.
As long as you pick sensible points to apply your fades and you don't more than a 1-2db at a time these changes should be transparent. The aim of the game here is to '˜equalize' the overall level of the mix and reduce its dynamic range so that any processors used later are not overloaded. You can generally get a good idea if you are done during a quick visual scan of the whole file.
The manually edited file is re-normalized.
Once you have sorted any peaks and checked the whole thing flows nicely, you can normalize again if you like to ensure you are squeezing as much level out of the recording as possible.
When deciding what processing to apply to a DJ mix, you only have to remember one thing: all of these tunes have already been mastered. You may have applied extra effects in the mix and changed some levels but they have most likely all been fed through a chain of pretty intense processing.
With this thought fresh in your head, it should become clear there is only one real way to approach this, very carefully. I tend to use a bus compressor but with just a tiny bit of gain reduction. 1-2db of reduction should be enough to capture any wayward peaks that were missed during the manual edit.
Some VERY light compression keeps things in check.
Other than this, I wouldn't advise you to apply any special effects, enhancers or anything else that might overly color your sound. The only thing I might use myself is a touch of equalization, we'll look at this next.
If you have recorded from vinyl or you used an analog signal path somewhere in your chain, you might experience a slight lack of high frequency at some point. It might be the case that some of the productions you have used were badly encoded, whatever the cause some mild EQ can go a long way to sorting out potential issues here.
I would suggest grabbing a linear phase EQ plug-in to retain absolute transparency and applying only a few dB of correction in the problem areas. An analyzer can help you to quickly home in on these sections.
A touch of analysis, some low cut and very small amount of high end boost.
If you find it's only a specific track or section of your mix that needs attention here you can use automation to introduce the corrective EQ. Alternatively, a more drastic option is to chop out the dull track, move it to another audio channel and treat it there.
Finish up with a dB or two of limiting to avoid overs.
Finally, you'll need to get your mix into a format that is right for delivery. If you plan to burn to CD or want to play the mix out in a live situation, then you are probably better to opt for an uncompressed format. If you are burning to CD, you will need to export your mix in 16 bit/44.1 and 24 bit exports can be used to play out directly from a computer.
High resolution export in action.
Remember that if you are moving from a higher resolution to a lower one, you will need to use dithering. This should only be applied once in the entire process and preferably at the end, so now is the time to do it.
If you are sending your mix out to a large audience online, or need to email a link to a specific person, then you will need a compressed format. You have a few to choose from here but of course MP3 is still the most popular. 320 kbit/s is probably the best choice as it's the highest bit rate MP3 will support. Lower resolutions such as 192 kbit/s will reduce file size and are often used for online playback.
MP3 export taking place.
Hopefully this has given you some pointers and will be some use to you next time you are preparing your latest mix for delivery. Happy mixing!