Saturday, February 9, 2013

Sibilance - How Singers and Audio Engineers deal with Sibilant Consonants

Let's talk about sibilance for a few minutes, since I briefly touched on it in my first Audio Recording Basics video on YouTube a few weeks ago.

Sibilance is another phenomena that generally makes a vocal performance less enjoyable. And by the way, sibilant consonants may sometimes also be referred to as stridents, obstacle fricatives, or obstacle affricates. But I don't think most singers or audio engineers need to memorize all of the different terms, as long as you know what sibilance means.

Basically, sibilance is the presence of certain "hissing" sounds in a singer's vocals. Generally, there are five letter combinations that can start a sound which leads to sibilance: S, Z, SH, CH, and J. Try vocalizing each of those sounds. You can probably hear/feel the hissing quite easily, right? Ok, try each one again for a second time, but slowly, and this time, think about something: the tongue is an incredibly versatile muscle. And a very fast one. When you vocal the S or the Z sound, think about the position that your tongue is in. The front of it is right up against the roof of your mouth, just behind your upper front teeth. It is mostly pressed up against the roof of your mouth with only a very small thin channel for air to flow through, which is why you hear the hissing sound, because the air comes out of that channel quite quickly. For the other three sounds, the SH and CH and J, the tongue is still up against the roof of the mouth, but a wider channel remains clear, which is why the air has more room to flow and the hissing is not quite as pronounced. Incidentally, I find it amazing how quickly a person's tongue moves during regular speech. I think a lot of people fail to appreciate how much work it does in the course of a conversation.

In terms of audio frequency, most sibilance occurs in the range from about 5k to 10k Hz. This is definitely the upper part of the range as far as vocals go. It's also interesting to note that as some people get old, they may suffer partially from a condition called presbycusis. This is basically a type of hearing loss, but it starts in upper frequencies. Basically, if presbycusis becomes advanced enough, the degraded ability to hear upper frequencies may creep down into the part of the spectrum that sibilance occupies, so the sibilance may seem to be less of a problem than it would have when the listener was younger.

Perhaps I shouldn't have titled this post to suggest that singers need to deal with this problem. The audio engineers play a much larger role in properly controlling sibilance in vocals, although it is good for singers to understand the phenomena. The first two things that I need to say about sibilance are that: (1) pop filters, which help deal with plosives, do not help reduce sibilance; and (2) microphone type and placement can make a huge difference.

I won't get into details about microphones here. The subject of microphone types and characteristics is incredibly complex. I want to put together a detailed tutorial video just about microphones, but to be honest, I don't even feel fully qualified to talk about them effectively, so I'll probably bring in an outside pro to help with that topic. But I can tell you a couple of brief points.

First, there are a several different types of microphones: dynamic, condenser, ribbon, crystal, and carbon. But the first two types are most common. Dynamic microphones are cost-effective, general-purpose microphones that are sturdy and robust. They can be used to record vocals, but would also be the type most often used to record various instruments, such as guitars (miking a guitar amp), etc. Condenser microphones are generally a bit higher quality, and are often better at capturing higher frequencies, but the drawbacks are that they are also a bit more fragile and they also need a small external power source (called phantom power) that usually runs to the microphone through the XLR signal cable attaching it to a mixing console. Often, the two types are mixed in recording sessions. For example, most instruments might be recorded with dynamic microphones, the vocalist with condensers, and the drum kit with a mix of several dynamic microphones capturing most of the kit with a couple of condenser mikes suspended overhead to capture a bit of extra high-end sizzle.

Anyway, the point of this background on microphones is not to tell you which one works best to reduce sibilance. The problem is that there is no specific answer to that. Different microphones (types OR models) can work more or less effectively, depending on the vocalist and to a less degree depending on the room. What works well with one vocalist might not be the best answer for the next vocalist.

Another interesting characteristic of microphones is that some of them are "directional." In other words, instead of picking up sounds equally well from all directions, there are certain directions from which sounds are recorded more or less easily. For example, in terms of recording fields, microphones can be classified as omni-directional, bi-directional, cardioid, super-cardioid, and hyper-cardioid. In other words, the microphone can record sounds differently depending on its orientation when it is set up. If you want to learn more about this topic, click here for a good post from DPA Microphones (warning, it's slightly technical).

The distance from the vocalist to the microphone should be greater than one might initially expect when trying to control sibilance. Depending on other factors, it might be common for the vocalist to place their mouth at least twelve inches away from the mike, and perhaps even eighteen inches away. It might also help for the vocalist to be above or below the usual horizontal plane. Of course, the easiest way to accomplish this is to adjust the angle of the microphone, not the singer.

So to sum up, if you're in a recording session and you're hearing the hissing of sibilance, your best options are as follows:
1. Try different microphones.
2. Make sure the vocalist's mouth is an appropriate distance from the microphone.
3. Angle the microphone slightly, according to what seems to work.

Other than that, there isn't a lot that can be done in terms of adjustments to the recording setup. I have occasionally heard people suggest that the vocalist can try chewing some gum and then sticking that gum up against the roof of their mouth. In some cases, this might help slightly, but this probably isn't a preferred approach because it would be annoying for the vocalist. Also, it would need to be a relatively small amount of gum, or else the singer is going to start to sound like they have something in their mouth, and his or her voice will start to sound different.

There is one other tool in the engineer's kit to reduce sibilance, which would occur during the post-recording period, when audio is being edited. There is a dynamic audio processing effect called "de-essing" which can use EQ'ing and compression to essentially reduce the volumes of certain frequencies within the bands in which sibilance is prevalent. Of course, it's always better to try to reduce problems at the recording stage, rather than hoping that a computer can resolve issues. If you want to learn a bit more about de-essing, click here to check out an article from Sound On Sound magazine.

Alright, hopefully this gives you some food for thought during your next recording session. Best of luck in your next project! And of course, if you'd like to check out some audio recording tutorial videos that I've put together on YouTube, click here to see a nicely organized index of the various tutorials that I've put together.

Follow Jonathan Clark on other sites:
        Main Site:
        Music Blog: