Hey! Since I’m still a newbie at dialogue editing I wanted to know how you approach finding a good EQ and tone for your dialogue.
Did you use reference dialogue when you started out? How do you determine how far you push the processing or is it a matter of “if it sounds good, don’t touch it”?
Do you avoid any tone-shaping until you hear it in the mix?
Appreciate you help!
Dialogue editing and processing is a DEEP rabbit hole. You can expect to spend a couple of years really feeling out your own philosophy and techniques. A really good starting point is John Purcell’s book.
Here’s my basic process – this assumes well-recorded VO that was cut flat with no comp or eq in the record chain:
- put a compressor and eq on the channel insert
- don’t apply any compression yet, but use the makeup gain to get the overall VO at about the loudness it’s going to be. For broadcast I shoot at -24 lkfs
- this is a super important first step, because loudness influences EQ choices. if you have the volume too soft you’ll make eq decisions that may not translate once the mix is coming together.
- sweep the area up around 1.5k to 6k and look for any unseemly peakiness. Many mics (even good high end ones) have some resonance there, and some voices exacerbate this. If you find something that really jumps out at you in this range, pull it back with eq. The more it jumps, the more you’ll have to pull back when you find it. If nothing jumps at you, don’t change anything.
- do the same in the low mids.
- this is another point where its important to do this step in this order. The reason is because those subtractive eq moves can GREATLY affect the dynamic range of the voice track as it goes into the compressor. IOW removing peaky eq points will reduce the amount that the compressor works later on.
- hipass until you hear it affect the voice. then back off again. The point of a hipass is only to eliminate non-voice crud. Its not to shape the low end.
- use a parametric to shape the low end to taste. this step is all about aesthetic. Don’t assume that you have to really do anything here – if the change isn’t additive to the sound, then don’t make it.
- use a shelf to shape the high end. same idea. don’t mix with your eyes on this step, just use your ears.
- de-ess to taste – this step comes AFTER shaping the high end.
- volume automate the track. Many talent will tend to tail off at the end of phrases, so lots of what this pass is for is to compensate for that. Note that we haven’t compressed anything yet.
- why not use another compressor here? Well, you can if you’re super pressed for time, but actually riding the fader across the entire track has the effect of allowing your aesthetic to really shine through. Your hand will make moves that a compressor will not. use it and trust it.
- route the track to a compressor on an aux. This is so that your automation moves play into the compressor – making it work less.
- set a super fast attack, and a reasonable release. move the threshold down until your compressing as much as you want to hear. in some contexts this is only 3db of compression or so. in other (louder ones) its much more.
- use the bus compressor’s makeup gain to restore the track to the right volume.
- de-ess a second time after the buss compressor if needed.
At this point you have a super tight, well balanced voice track. mix the music and fx in against this track to taste.