Podcast: Play in new window | Download (Duration: 8:14 — 7.9MB) | Embed

Subscribe: Spotify | TuneIn | RSS

If you’ve got aspirations to play on a big club or festival sound system one day, you’ve got to tame harsh frequencies in your tracks or you’re gonna people’s ears bleed.   Today we’ll dive into some techniques to reduce harshness in your mixes.

You’ll Learn

• Causes of harshness in a mix
• Some tricks for reducing harshness
• Plugins I like for de-harshing a mix

Resources

• 8 Tips for Taming Harsh Treble in the Mix
• Five ways to De-harsh Your Mixes
• iZotope Track Assistant
• PLP 022: Mixing a Track with iZotope Neutron 3 Artificial Intelligence
• iZotope Ozone 8 and Neutron 2 Bundle Review

Transcript

What’s up Heroes, welcome to the Producer Life Podcast Episode 34.  If you’ve got aspirations to play on a big club or festival sound system one day, you’ve got to tame harsh frequencies in your tracks or you’re gonna people’s ears bleed.   

But first, cue the intro music.

What do we mean by harsh sounds? The human ear is particularly attuned to the frequencies where human speech occurs, especially the 2-8khz. This harshness can be caused by a lot of things such as: boosting higher frequencies too much, harsh sounding synths, multiple sounds all piling up on top of one another, too much compression or distortion,  or even poor quality mics or preamps. 

 In order to hear those harsh sounds, you’re going to want a couple of things. First, a good reference track for comparison in the genre you’re producing in. Second, fresh ears. Give your track a rest for a day or two so you’re listening to it fresh. Third, some tools to work with.

Start by thinking about the purpose of your track within the context of the mix.  If it’s going to be buried in the mix, you can probably throw a high-pass filter on it and call it a day. However, if it’s one of your front-and-center sounds, like a vocal or lead synth… you want to spend a lot more time on it.   

We’re gonna use this synth loop sample from Loopcloud as an example. 

Idea one: Comb out the harsh frequencies with sharp notched EQ. Here’s the process: start with an EQ, like EQ8 in Ableton. Create a narrow but heavy boost (15db works well with 3.5Q, which is the bandwidth of a particular filter. 3.5 is pretty narrow. Turn your volume way up on your headphones and slowly sweep through the 2-8k range listening for particularly harsh tones or ringing. Putting EQ8 in audition mode may also help, just click on the headphones on the right hand side of the device and then left click on the filter dot and drag it to the right so you’re only hearing the frequency boost. Only do this for a short period of time or you’ll damage your hearing. It sounds something like this, only much louder.

When you find a spot that has a particularly harsh resonance, cut that sound by 2-4db.  Keep in mind if the Q (narrowness of the notch) is too sharp, it can give a very unnatural tone. Also, it’s a good idea to do this BEFORE the compressor in your channel strip because compressors don’t effect all frequencies the same, and while you’re still doing the mixdown. Pay particular attention to sharp metallic sounds like cymbals reverse crashes, risers.   Our revised track sounds like this with a 4db cut at about 3k. Here’s before….  Here’s after.

Sometimes, a regular EQ like EQ 8 in Ableton isn’t enough…. you need a dynamic EQ which acts as a combination between a compressor and an EQ. A dynamic EQ only reduces a sound when it meets a certain threshold, so for harsh sounds in a vocal, for example, where the singer may only have harshness in a certain verse or refrain, this can be an ideal solution.   I happen to like iZotope products and their EQ, but there are lots of options out there. 

Now, after notching out harsh frequencies, you may find your track sounds lifeless or hollow. If so, try adding an exciter or saturator to the channel after the EQ notch, then add back in a boost of the same amount after that. For example, your lead synth might be harsh at 4500hz. You notch 4db out, then add a little saturation, then boost 4500hz at 4db with the same EQ curve.  Sweep the frequencies again to ensure you didn’t create additional problems.

Another possibility: our ears perceive frequencies in relation to one another. For example, if you want a mix to sound brighter, sometimes cutting the lows is more effective than boosting the highs. Soooo…. When you’re thinking about harsh higher frequencies you may notch out, consider also providing a slight bump in the lower frequencies to balance that out.

Another option is to grab a specialized plugin designed to address these issues.  There are a variety of tools out there that can help tame harsh tones. Izotope’s Neutron 3 offers an AI-based track assistant that can give you a good place to start with your EQing and can help you make choices about which frequencies to notch. In this case I ran our sample through Neutron 3 and put it on it’s “balanced” settings and used auto-detect for the instrument. It listened for a minute, and then recommended some initial EQ curves among other things.  Unfortunately, it suggested a steep dynamic EQ at 400hz and a boost around 1k… so that wasn’t really what I was looking for. Use your eyes and ears. If it doesn’t make sense, discard it. 

Izotope’s Spectral Shaper is another useful tool. Spectral Shaper is a multiband compressor with a “tone” slider that lets you adjust how much brightness you want in certain bands. Soothe2 is another one which is designed specifically to address harshness in a mix and has been well reviewed online.  I’ve tried Sooth’s free trial, but not enough to have a strong opinion.   

Another possibility is that you’re using too many plugins which is causing a buildup of frequencies in a painful zone. If you’ve got more than three plugins on a channel, try muting all of them and then adding them back in one at a time and seeing if that’s causing the painful frequencies. 

Sometimes, the harshness comes from painful sibilance that is caused when our tongue directs a stream of air towards the teeth to form the S sound. This can cause a harsh, sharp sound at 8khz.  There are a bunch of Dessers on the market, including the stock one in ableton, that can help address this particular problem.  Because De-Essers use dynamic processing, meaning they respond to changes in the track over time, they can also be good for other instruments such as harsh cymbals or even guitars. Most De-essers have an option to allow you to listen to only the frequencies they’re effecting, sometimes called the delta or audition button. Make sure you’re using that to gently tweak those sounds. 

Keep in mind that any of these techniques could also be used on an aux or bus track (a group of tracks). For example, if you’ve got multiple vocals layered in the chorus, you might try one of these techniques on the vocal bus instead of each individual track.

So, what do you do to reduce harshness in your mixes? I’d love to hear your tricks, just chime in at producerlifepodcast.com episode 34 or over in our private Facebook group at https://www.facebook.com/groups/ProducerLifePodcast 

Until next week, this is the House Ninja reminding you to be somebody’s Hero today.