How to EQ Your Dialogue from Start to Finish: Mastering Step 8

So, you've worked through the recording process and editing of your voice recording.

Maybe it's sermon audio from the past week or your podcast.

It's starting to sound really good after you've cleaned things up in the editing phase by dealing with background noise, stopping sibilance and removing resonances.

Now we enter the mastering phase to learn how to use EQ in this process of finalizing audio before it's released to your listeners.

Firstly, how do we use EQ in mastering to correct or shore up any last minute problem areas that may surface?

Finally, in the next article we'll learn how to use EQ in mastering to balance our voice recording's overall sound before it hits the limiter.

Is There Actually A Problem?

Ideally, you've dealt with the majority of the EQ problems by the time you've reached the mastering stage.

However, problem spots will still crop up - perhaps they weren't fully dealt with in editing or slipped by unnoticed. No need to worry.

But how do you know for sure if there's a problem for corrective EQ in the mastering phase?

The best way to unearth last minute EQ issues will involve careful listening and comparison with a good sounding reference.

Some issues might be very obvious, like a persistent stabbing sibilant area or a stubborn resonance popping out.

Other issues might not be apparent until you do some comparative listening with your reference voice recording...

For example, issues like needing a broad boost in the upper midrange for clarity or a cut in the lower mids to relieve some muddiness and to let the higher frequencies shine through.

If possible in your audio software, try setting up your reference in a separate track from the track you're mastering for comparison.

Make sure to match the volume between them to be able to make accurate judgment calls on any problem areas you hear.

Listen carefully and apply any corrective EQ that's needed.

Another method is to get your voice recording sounding as good as you can in the mastering session.

Then process some test master versions and compare those with your good sounding reference (always matching the volume of course!).

Finally, take note of any EQ changes that need to be made to your audio back in your mastering session.

Lastly, no matter what method you use, try to do your comparison listening with several different speakers.

For instance, use your best headphones, next some speakers you're familiar with and finally listen on your phone.

This might sound tedious, but the different listening sources will give you the best "picture" of what your voice recording actually sounds like.

Problem areas that can be corrected with EQ will become more apparent using this method.

For instance, oftentimes I can best gauge if I need to de-ess some more when listening on my phone (it has a terrible sounding speaker, but helps me focus on those upper midrange areas).

Think of each listening source as another piece of the overall puzzle of what your dialogue actually sounds like.

Where Do You Put The Corrective EQ?

After you've noticed problems or compiled notes from your listening comparison, you'll place your EQ in your mastering chain. But where?

Putting the corrective EQ near the beginning of the mastering chain makes a lot of sense.

This is because you generally want to shore up and remove any remaining problem areas before they're processed through the mastering chain.

The idea is that you want to start the mastering process with your voice recording being as free of problems as possible.

This is before your audio travels through other processing down the mastering chain (e.g. compressors, color EQ for balancing and limiters).

With this method of correcting first, you allow yourself to focus on the main goals of voice recording mastering:

• balancing the overall tone...
• sweetening the tone...
• and bringing up the loudness to proper levels for playback.

Take a look at the image of my voice-over project's mastering chain below.

Notice that I put my corrective EQ, the FabFilter Pro Q3, near the beginning of the chain:

• just after Ableton Live's gain device...
• but before the compressor (Vertigo VSC-2)...
• then a color EQ for balancing the tone (Dangerous BAX EQ)...
• and finally the limiter:

You can see that I'm using FabFilter's EQ, but any high quality parametric EQ will work (like the stock parametric EQ in Ableton Live or the FREE TDR Nova).

How Do You Use Corrective EQ?

So how do you translate the problem areas you were hearing (during your comparison to your reference) into corrective EQ settings?

Here was some of my thinking based off of my notes that may guide your EQ moves as well:

• I knew my voice recording was missing some of the important upper midrange clarity that the reference had...
  
  
• That meant I would need to do some fishing around with a small broad boost in the upper midrange (2 to 4 kHz) to see what would bring out that clarity I wanted to hear...
  
  
• I also knew that I couldn't do too much boosting there (2 to 4 kHz), because of the remaining sibilance in the upper midrange that was still plaguing my recording...
  
  
• That meant I could try some reduction to the opposite area of the upper midrange (the lower midrange: 250 to 500 Hz)...
  
  
• A reduction down there would lessen a boxy resonance I was hearing and at the same time unveil the upper midrange a bit (reduce some low to bring out the high)...
  
  
• That small cut in the lower midrange together with the small boost above made a significant difference in pushing my tone closer to my reference...
  
  
• But we can't forget the pesky sibilant areas - with the small boost in that area, I had to be sure to de-ess those small, spiking areas with cuts and dynamic EQ.

Here are my settings on the FabFilter Pro Q3 (see images below) to understand how my thinking above translated into specific EQ moves:

• Low Cut: 20.00 Hz - frequency; 1.0 - Q; 18 dB/octave - slope steepness
  
  
• Band 2 (bell): 445.00 Hz - frequency (low midrange cut)
  -0.50 dB - amount (very small amount)
  1.5 - Q (width is set a little tighter on the resonant area)
  
  
  
• Band 3 (bell): 3442.4 Hz - frequency (upper midrange boost)
  +0.75 dB - amount (small amount)
  1.0 - Q (width is broad to cover general area)
  
  
  
• Band 4 (bell w/dynamic EQ): 3979.7 Hz (sibilance)
  -1.00 dB
  12.10 - Q
  
  
• Band 5 (bell w/dynamic EQ): 4610.8 Hz (sibilance)
  -1.00 dB
  12.38 - Q
  
  
• Band 6 (bell w/dynamic EQ): 5128.0 Hz (sibilance)
  -1.00 dB
  12.63 - Q
  
  
• Band 7 (bell w/dynamic EQ): 9240.1 Hz (sibilance)
  0.00 dB
  13.75 - Q
  
  
• Band 8 (bell w/dynamic EQ): 10793 Hz (sibilance)
  0.00 dB
  13.74 - Q
  
  
• High Cut: 30,000 Hz - frequency; 1.0 - Q; 18 dB/octave - slope steepness

If you would like to grow in learning to convert problems in your voice recordings to simple audio corrections, grab this valuable, FREE resource: "Voice Recording Problem Solving Guide".

Let's Recap

In review, the best way to find any remaining problems in your voice recording is to do a careful comparison with your good sounding reference recording.

Even try this listening on several different sets of speakers or headphones.

These different listening perspectives can offer helpful clues about your overall sound and the steps to take with your EQ corrections in the mastering phase. 

Next, remember to place your parametric EQ near the beginning of your mastering chain.

This is to remove any problems you find before they travel through the other plug-ins and negatively affect the final tone of your dialogue's master.

Finally, begin your EQ moves by addressing problems that popped out or discrepancies you noticed from your listening comparison.

Try small cuts or boosts with broad bell filters in specific, targeted areas to correct tonal differences to your reference.

Remember that small cuts and boosts of just ±0.5 dB can add up to a significant change in your overall sound.

Have YOU used these steps on a recent voice recording to improve the sound during mastering? How did they help?

Would love to hear your thoughts in a comment below!

Cheers, Bryan
Your Dialogue Editing Mentor