How to Reduce/Eliminate Sibilance from Harsh Vocals

RangeOfSounds.com is reader-supported. We may earn a small commission through products purchased using links on this page.

A general EQ can usually fix your vocalist’s pitch, but this isn’t always the case. Along with pitch, sibilance can cause some harsh high frequencies that may be uneasy on the ears.

One thing to consider is that no two vocalists or microphones are the same and because EQ, compression, and De”S” worked with one vocalist, doesn’t mean it will work again in the same way. Approach each performer’s voice like a unique project of its own. Don’t assume past practices will work with this new vocalist.

Let’s also remember that sibilance and harsh vocals do not always go hand in hand. Sibilance is created with strongly spoken consonants that produce air from the lips and tongue. Harsh vocals can be created without sibilance and can be associated with the pitch of a vocalist.

Definitions

Sibilant is defined as the following:

sibilant [ sib-uh-luh nt ]

Adjective

  1. Hissing
  2. Phonetics. characterized by a hissing sound; noting sounds like those spelled with s in this [th is] /ðɪs/, rose [rohz] /roʊz/, pressure [ presh -er] /ˈprɛʃ ər/, pleasure [ plezh -er]/ˈplɛʒ ər/, and certain similar uses of ch, sh, z, zh, etc.

Harsh is defined as follows:

Harsh [ härSH/ ]

Adjective

  1. Unpleasantly rough or jarring to the senses

Approaching Harsh Vocals with EQ

Let’s start with harsh vocals. Let’s assume we’re not dealing with sibilance but working with a vocalist that has a wide vocal range. Let’s say we’re dealing with someone like Dryden Mitchell, Chester Bennington, or even MJ himself (R.I.P. to the gods). Our artist sings at velocities that scream and their pitches can make an average soprano look bad.

One thing to consider is mic placement. Tilt the mic a few degrees away from the source, off axis, either left or right or to the back.

Have your source 6 to 10” from the mic. These are some steps to practice.

Let’s say we had no option in which the mic or performer were positioned but are engineering post. Eq is key. In some cases, I’ve placed low pass filters at 2500 Hz and had to eliminate the high end entirely. I know this may sound unheard of, but you should have heard the source. The performer had so much range that eliminating the high frequencies didn’t affect the final track in a way that would be audible to the average listener. It may have been noticeable to an engineer, but the final product was still beautiful. Without the aggressive EQ, the voice seemed to continue to distort with levels of -15 to -20db. I’m not saying to EQ all of your vocalists like this; I’m just giving you an extreme example.

Let’s say this isn’t the case, but at least you know what can be done. Let’s say our vocals sound decent but is still a little harsher than we’d like.

Remember EQ is key. With an EQ, you can create a narrow bandwidth and sweep across the EQ to find the harsh places in the vocals. Make the bandwidth as narrow as you can. Be careful while doing this, as it can be very hard on your ears. There are some DAWs that have an inverse frequency EQ, where you can invert the other frequencies to be nonexistent and work with the desired frequency without turning it up. With Protools on PC, if you press SHIFT + WINDOWS KEY while adjusting any knob, it will solo the frequency. While sweeping, you’ll hear the frequencies that are unpleasant and they’ll become even more unpleasant. Some say these frequencies sound like feedback and bite the ear. In these frequencies, keep your bandwidth tight and turn them down. Here is an example of sweeping with the 5th band. Remember, while sweeping across the EQ, the track must be playing so you can hear the frequencies that need adjusting. Make sure the whole mix is playing because frequencies that may sound harsh soloed may fit just fine in the full mix. The 6th band is an example of where we isolated a frequency that needed to be turned down.

Now, you can do this with several bands. Sweep, find the harsh spots, tones, frequencies and reduce. Remember, all vocalists and microphones are different and adjust accordingly.

So, we’ve approached harsh vocals, but what about harsh sibilance?

Approaching Harsh Sibilance with a De-esser

Remember, sibilance is when a performer is pronouncing harsh consonants. Of course, mic placement helps, especially placing your mic off axis by a few degrees, but what if the sibilance is still affecting your final take. What do you do then?

There are a few different ways to tackle this sibilance issue, and we’ll start with a commonly used tool in today’s digital production world, the de-eser. What does a de-esser do, you might ask? A de-esser compresses the harsh consonants/sibilance in the track.

Let’s give you an example of sibilance: “Sammy bought seashells by the sea shore.”

S-ammy bought S-ea-Sh-ell-S by the S-ea Sh-ore.

When the S’s are enunciated, the vocals take can and sometimes will sound harsh.

A de-esser can help you isolate the high frequencies where those consonants are striking and compresses those frequencies so they become less harsh. The de-esser is a great automated tool to help with sibilance issues.

De-essers typically have the following functions:

  • Detection/Suppressor
  • Sensitivity
  • Strength
  • Smoothing

You’ll need to find the frequencies that are affecting the mix the most, often ranging from 3k to 9k. Let’s say the frequency being affected the most in this instance is the 7.5k. Now you can begin adjusting sensitivity. The sensitivity is the threshold you’d find in any compressor.

You’ll want to find the sweet spot in between compressing too much and not compressing enough. With a de-esser, you can make your consonants disappear entirely, so you must use with caution.

When using a de-esser more than you should, the singer will begin to sound slurred or begin to sound as if they had a lisp.

The strength then adjusts how much of the chosen frequency you would like attenuated or reduced. Last but not least is smoothing. Smoothing is the attack time you would like the de-esser to perform its compression.

Don’t forget EQ can help a lot with sibilance issues. Make sure you EQ first, and try to eliminate the harsh frequencies before running an automated compressor like a de-esser. It is recommended to EQ and compress before using a de-esser.

In Conclusion:

  1. Start with EQ and Compression
  2. solate harsh frequencies
  3. Adjust threshold of de-esser
  4. Adjust strength; apply too much and back off gently until natural sounding 5. Adjust smoothing or attack time; quicker attack will reduce harsh sibilance sooner