Microphone Technique: how to reduce sibilance when recording
Vocal sibilance can be pretty horrible to listen to. It’s harsh, unpleasant, and really jumps out. Luckily it’s easy to get rid of when recording, or in post.
In this article we’ll tell you where sibilant frequencies appear, give you advice on reducing sibilance at the recording stage, answer the question “what does sibilance sound like,'' and nail exactly: what is sibilance in audio?
We’ll also illuminate whether a pop filter for sibilance actually does anything, and how you can combat vocal sibilance in your editing software.
Sibilance occurs in recordings when a person speaking into a microphone says an “ess” or “tee” sound. These syllables are said to be syballent because they contain a lot of energy which makes them jump out in audio recordings.
The reason sibilant syllables contain so much energy is that they contain information high up the audio frequency spectrum. We’ll jump into the specific frequency range of syballant syllables in the next section.
The above waveform is a recording of a person saying the word “silly”. You can see where the sibilance occurs in the first half of the word. The waveform is much tighter showing a higher frequency. A higher frequency, by definition, indicates that more vibrations are happening every second. Quicker vibrations means more energy which is why sibilance can sound so loud and harsh.
What frequency causes sibilance?
Sibilance occurs in the 5 to 10 kHz frequency range. Below you can see a screenshot of a spectrum analyser as the recording from the previous section of a person saying “silly” is played.
This screenshot shows the moment the sibilant first syllable of the word is spoken i.e “si”. As you can see there is a lot of energy in the 5kHz to 10 kHz range with a notable peak at around 5500Hz.
This next image shows the moment the second non-sibilant syllable of the word is played i.e “lly”. As you can see there is no energy above 5k, and thus no sibilance is occurring.
If you have a sibilant voice recording and you want to reduce sibilance then fortunately it’s very easy to do. There are a few methods of reducing sibilance, here we’ll run you through some of the classic, old school methods, and then show you the absolute best and easiest way.
A quick way of reducing sibilance is to EQ out the harsh frequencies in the 5kHz to 10kHz range. This is an easy fix, however removing these frequencies can alter the tone of your recording when no sibilance is occurring, so it’s not recommended.
Do it by hand
One effective way to reduce sibilance is to cut out each instance of sibilance in your recording and manually turn down the gain of the sibilance syllable. You can see an example of this with the “silly” waveform we showed earlier.
Here we’ve reduced the sibilant syllable in gain by about 8dB.
This is an effective and accurate way to get rid of harsh s sounds in your sibilant voice recordings. However, the big downside of this method is that it is very time consuming. If you’re working with a long recording, trawling through and cutting out every instance of sibilance will take a long, long time. For both quick and effective sibilance reduction you’ll want to use a de-esser.
A de-esser is a specialised tool designed to remove vocal sibilance from your recordings. One of the best, and easiest to use de-essers out there is ERA De-esser. It’s a simple one knob plugin that uses an intelligent algorithm to only remove sibilant frequencies from your sounds. Simply turn up the knob until the sibilant frequencies in your recording are gone.
Let’s continue the example we’ve been using so far in this article. Here is a recording of the word “silly” being spoken.
Notice how the first syllable of the word (i.e the “si”) is very harsh. It doesn’t sound like how the word would sound if someone was just talking to you in person – the ess is emphasised. This is sibilance.
Here is what the “si” on its own sounds like, just so you can hear exactly what we’re listening for.
It is very hard to entirely remove sibilance from your recordings using microphone technique. That said, there are several things you can do to your mic to reduce unwanted pops and sibilance when recording.
In the next few sections we’ll tell you how to reduce sibilance with mic technique, explain how pop filters work with sibilance, and discuss whether you can reduce speaker sibilance.
How do I lower my microphone sibilance?
Here are a few techniques you can use to lower the sibilance in recordings with microphone technique.
Distance yourself from the mic
The reason why sibilance is so bad in recordings, but you don’t really notice it in day to day conversations, is that when you record a voice you are much closer to the microphone than you normally are to someone you are talking to. Because of this, the energy in these high frequency sibilant syllables doesn’t have time to dissipate.
Therefore, if you want to reduce sibilance in your voice recordings, distancing yourself further from the microphone can help.
Tilt the mic
As a general rule, microphones are more directional at higher frequencies, and less directional at lower frequencies. This means that if you tilt your microphone about 15 degrees away from your mouth, higher frequencies like sibilance will be quieter in your recordings whereas lower, more desirable frequencies will be clear.
A good rule of thumb here is (if the mic is six inches away from you) to point the microphone towards your neck.
Does a pop filter help with sibilance?
Pop filters don’t help to reduce sibilance. They’re designed to tackle plosives which are a similar, but different problem you may encounter when recording.
Plosives occur when you say Ps and Bs. A burst of air rushes from your mouth, hits your microphone diaphragm, and is turned into a bass signal in your recording. Pop filters dissipate this burst of air and in doing so reduce the problem.
Sibilance isn’t caused by a burst of air, it occurs as a result of high energy, high frequency audio. Therefore pop filters don’t do anything to help with sibilance.
How can I reduce speaker sibilance?
There aren’t any techniques you can use to reduce sibilance in your speakers. Instead you should use the methods discussed earlier in this article to reduce sibilance at the recording and editing stages of your audio production.
Harsh audio is normally a result of too much high frequency information in your recording. Fortunately this is very easy to fix with a basic EQ.
Most video and audio editors come with a built in EQ. Here we’ll demonstrate with Premiere Pro’s parametric EQ.
All you need to do to soften harsh audio is apply a high shelf cut with an EQ. This means you effectively turn down the volume of all audio content above a certain frequency.
Rolling the audio off around 6kHz will work, but experiment to see what sounds right. When it comes to how much you should cut you want to reduce the harshness without making your audio dull. Again, experimentation is key here.
With the info in this article, sibilance problems in your audio should be a thing of the past. If you want to learn more about sibilance and dealing with it in your recordings, we have a few more guides on our blog. We also have detailed instructions on using a de-esser and a list with the Best Microphones for Recording to help you choose wisely the microphone type for your recording.
If you’re looking for quick, easy and very effective ways to combat problems in your audio recordings then check out the Accusonus ERA Bundle. It’s a toolbox of plugins designed to deal with all common issues sound recorders face.