Learn Music and Audio Production | iZotope Tips and Tutorials
When to Use Dynamic EQ In a Mix
Using EQ in mixing is pretty straightforward, but your biggest challenge will often be sculpting audio that varies in frequency content over time. A cut to tame harsh vocal frequencies might work for one phrase, but compromise intelligibility in the next. Rolling out lows on a drum kit might reduce masking when the bass is present in the chorus, but might make the drums sound thin when the bass isn’t present. You can try automation to connect the dots, but that’s both time-consuming and unnatural for these types of moves. Enter dynamic EQ.
This article references a previous version of Neutron. Learn about Neutron 4 and its powerful features including Assistant View, Target Library, Unmask, and more by clicking here.
What is dynamic EQ?
Dynamic EQ has the three EQ parameters you’re used to—frequency, gain, and Q—with the addition of a threshold like you're used to seeing on compressors. When audio exceeds the threshold in a specific frequency band, it triggers either a boost or cut of the EQ filter. This combination allows you to use EQ in a way that responds and adapts to incoming audio, only affecting it when that audio crosses the threshold.
In this article, we’ll look at some of the corrective and creative possibilities of dynamic EQ.
Generally speaking, why use a dynamic EQ?
Most EQ filters are static, meaning they apply their boost or cut at all times. While it's unlikely that static EQ filters will lose their place in the world of audio any time soon, dynamic EQs present an alternative for producers and engineers looking to process audio transparently by only affecting it when needed. Check out the following video to get an idea what we can do with dynamic EQ, specifically in Neutron:
Let’s look at some of the most common applications.
1. Use dynamic EQ to tame momentary resonances
This is probably the most common use for dynamic EQ. While a vocal or instrument performance may be pretty smooth overall, there might be moments where a certain frequency pokes out of the mix in a distracting way. Since that frequency only needs to be attenuated when it gets too loud, we can use dynamic EQ to tame this in a transparent way.
This same problem happens in all kinds of audio signals, particularly of the live variety. Not only do you have to deal with resonances caused by the instrument, but now there are room acoustics to consider. Room acoustics can reinforce resonances that already exist, and even introduce new ones depending on your recording space and sound source.
Instead of spending your afternoon hunting for frequencies, you can use the Learn feature in the Neutron EQ module to locate them, then hit “Dynamic Mode” to set a threshold for each node.
You can check out how the Learn feature works in the video below, which works the same for all version of Neutron:
2. Use dynamic EQ with a sidechain input to unmask instruments from each other
Masking issues present themselves between plenty of instrument combos, like vocals and backgrounds, vocals and guitars, and bass and mid-range synths. Because these pairs of instruments share characteristic frequencies, these instruments compete for the listener’s attention, potentially causing a lack of clarity and a muddy mix.
The obvious choice is to carve space out for both instruments with static EQs, but masking may only occur during certain moments in the track. If this is the case, or if we’re just looking to unmask our instruments subtly, we can turn to dynamic EQ.
For example, let’s say we have a vocal that’s getting masked by some guitars, especially during the hook line in the vocal. We don’t want the guitars to totally drop in volume during this line, but we might want to attenuate some guitar frequencies that are important to the vocal, letting the vocal shine through more during an important lyrical moment.
You can do this pretty easily with Neutron. First we drop an instance of Neutron on the guitar channel and pull up the EQ module. Next, we determine the frequencies where the most masking is occurring (you could use the Masking Meter feature to find this). After that, we place an EQ node on the guitar EQ and activate both Dynamic and Sidechain mode (in the “Ext” setting).
Then, we can create a return track and send the vocal to that channel at 0 dB, so the signal leaving the return track is as loud as the actual vocal track. We can route audio from this return track to the instance of Neutron on the guitar channel, and therefore use the vocal signal from the return track as the sidechain input. If we set the threshold appropriately, the important vocal frequencies will drop in the guitar track when the vocal gets loud enough.
The application described above applies to these instrument combinations and many others too—the main vocal may clash with the backgrounds, or sustained notes in the bassline may aggravate the lower parts of a synth.
Using a node or two and the sidechain trigger in the Dynamic section of Neutron’s EQ will allow you to get better control of these brief moments of collision so your mixes sound cohesive and free of distraction—without needing to make compromises with a static EQ.
3. Use dynamic EQ to update the classic kick-bass sidechain
Perhaps the most infamous masking example happens between the kick drum and bass track.
The most common solution is to use the kick as a sidechain trigger to temporarily duck the bass with broadband compression every time it hits. While this works, it requires you to duck the entire bass and not just the frequencies that interfere with the kick. This can produce an obvious pumping effect that isn’t always desired (though for some genres, it may be what you’re after stylistically).
Dynamic EQ allows you to zero in on the specific frequency band causing the fuss between the kick and bass. Again, you can use the sidechain workflow in the Neutron EQ to perform this unmasking:
1. Add an instance of Neutron on the bass track, activate Dynamic and Sidechain modes on a node, and select “Ext” as the sidechain source.
2. Create a return track and send the kick to it.
3. Send the audio from the return track into the instance of Neutron on the bass track.
4. The kick will now trigger the bass sidechain, so you can drag the node to wherever masking occurs and it will duck this band according to the threshold parameter.
4. Use dynamic EQ to get drums to “talk” to each other
Let’s imagine a snare in a drum submix is struggling to cut through because it is being masked by some of the lower frequencies of the hi-hats. If you apply a broad static EQ cut around 5–6 kHz to the entire submix you’ll end up dulling the snare sound. You can also try a cut on the hi-hat sound itself, but the masking issue only arises when the snare is present so this isn’t really helpful.
With the dynamic EQ in Neutron, you can use the snare hit as a sidechain input to the hi-hat track so the conflicting frequencies are only cut when the snare hits. In case we didn’t have all the stems, we could even do this with an instance of Neutron on the submix itself, using the Internal Sidechain mode (Int).
This is all laid out in the image below—Node 3 is positioned over the snare fundamental and Node 7 is sitting on important hi-hat information. I selected Node 3 as the internal sidechain trigger for Node 7, meaning Node 7 will be attenuated when Node 3 crosses the threshold of -50.0 dB.
Note that so far in this article, we’ve set all dynamic EQ nodes to attenuate audio, but we can certainly boost with dynamic EQ too, and even in a sidechaining context. If I wanted to boost these hi-hate frequencies every time the snare landed, I would just select “Up” instead of “Down” in the Neutron EQ node.
5. Use dynamic EQ to pinpoint vocal resonances
Some of the trickiest issues to resolve in a mix are the sporadic resonant peaks inherent to certain vocal ranges, or certain vocal sounds like sibilance. Sometimes these issues are so pervasive you can see them poking through on an EQ graph, but they can also blend into a performance in sneakier ways.
With a static EQ, you might start making small cuts across the spectrum to attenuate resonances as they happen in a performance. In a particularly resonant vocal, this approach can leave you with several cuts that, while solving the issue when those resonances are present, pull away important vocal content when the resonances aren’t there.
The problem here may also be twofold. First, we have the issue of these static cuts being applied all the time, not just when the resonances are present. We can apply some dynamic EQ like we’ve done in other examples above to solve this problem.
However, the vocalist may also just have a resonant harmonic in their voice, which follows their pitch, and needs to be attenuated at different frequencies. Unless you automated the frequency of a dynamic EQ node to follow that harmonic—which would be incredibly tedious—a typical dynamic EQ may not solve the problem.
This is where the Follow EQ feature in the Nectar EQ module can be helpful. This is similar to dynamic EQ, because processing adapts to the incoming audio, but Follow EQ causes an EQ node to follow the frequency of a moving harmonic. This way, instead of drawing tons of automation, you can just let Nectar do the work for you.
If you are having trouble finding your vocal’s most resonant frequencies, run a pass of Vocal Assistant, which creates a custom preset for your vocal based on its characteristics, and head to the first EQ module in the chain. A series of nodes will be placed at various points across the spectrum, indicating where possible issues lie.
Bonus: use Follow EQ to clean up mud under a vocal
Okay, so Follow EQ isn’t exactly dynamic EQ, but we had to include one more way that it can be a lifesaver:
Below 100 Hz, vocals carry little important information, making it standard practice to roll off everything below this point with a high-pass filter near the fundamental. You don’t want to thin a vocal out, but you do want to remove rumble and other unnecessary noise.
Using Follow EQ in Nectar, you can introduce a high-pass filter that follows the first fundamental as it moves from pitch to pitch, taking the guesswork out of choosing a cutoff point.
This is helpful for all vocal performances, but particularly for those that are erratic in nature. Even if the fundamental jumps multiple octaves, the high-pass node will follow it closely so you don’t have to draw in filter automation or slice out certain phrases to new tracks for different EQ processing.
Dynamic EQ is one of the more versatile types of processing out there, and from sculpting vocals to dulling resonances, and unmasking similar-sounding instruments, it should now be clear how using it can greatly improve a mix when needed.
For those of you who were unfamiliar with dynamic EQ before this article, follow the tips listed and you will surely find new ways to employ it for both creative and corrective results.