Recording Tips: Limiters

If you have spent any amount of time recording and mixing, you have likely tried to keep a sound from clipping or tried to boost the overall volume of a recorded sound only to have it clip because of a small section or a single sound that tends to be much louder. We’ve already covered two types of compression, basic and side-chain, now I’d like to discuss limiters, which will help with the above issues.

A limiter is a compressor that is typically set to a higher ratio and faster attack (see compressor basics). The goal of a limiter is to “limit” a sound to a specific top volume. Limiting effectively reduces the dynamic range of a sound, this means that if set correctly, it will turn down the louder sections to bring their volume closer to the softer sections. Limiters are used frequently in live sound applications to prevent sounds that are too loud from damaging speakers. In a studio setting limiters are often used as a tool to reduce louder sounds so that the volume of the track as a whole can be increased without clipping or distorting. We will focus here on the studio usage only.

A compressor begins limiting once the ratio is set above 5:1 although it is usually a higher ratio setting, closer to 10:1. Also, attack times are generally much faster. To set up a limiter, start by setting your threshold (or ceiling) to the dB level that you want the sound to start to being effected. An example of this would be if you had a sound recorded where the majority of the sound was below -10 dB, with some areas that were around -2 dB and you wanted the volumes of those sections to be closer to the rest of the mix, you would set your threshold somewhere below -2 dB, but probably not more than -10 dB. Next, adjust your ratio to the desired amount of effect you want to use; 10:1 is a good starting point. Now you will use a fairly quick attack setting (this will be a “to taste” adjustment) to tell the limiter how quickly you want the effect to be executed on the sound. Finally, you can use a make-up gain to increase the overall sound of the newly limited track, which should provide for an overall higher volume without clipping. If you limit too much and use an attack time that is too fast, the result will likely be undesirable.

Limiting is a great way to protect speakers, smooth out performances and increase volume. Remember there is no standard setting for any two sounds, so spend some time tweaking the parameters to get the best effect possible. Try not to use more limiting than is necessary as the goal is to improve the overall mix without being overly obvious. The untrained ear should have no idea that it is there at all (unless that’s desired goal).

Recording Tips: Side Chain Compression

Now that we have covered the basics on both the compressor and the EQ, we can now start discussing a few of the cool ways to use these tools to improve our recordings. For this discussion, I’ll be discussing a very useful compressor function for those working with voice; Side chain compression.

Side chain compression is feeding one audio source into the compressor to trigger the compression of another audio source. While this can be used in many different ways, I will only focus on its basic use for voice. The side chain compressor is most frequently used for “ducking” or turning down a sound to get out of the way of another sound. You might use this if you are mixing a music bed underneath a voice over track. If set up to do so, this will allow the music to be automatically turned down when the voice track is running and swell back up once the track is silent. Some compressors have this built in and can simply be switched to side chain mode, but it is more likely that you will have to seek one of these out. If you do not have one, there are some good free or very cheap ones that can be downloaded with very little searching. Please realize that every compressor is a little different in its functionality, in this example I’m using a basic Digirack Compressor/Limiter. I will explain how to use a side chain compressor to “duck” a music bed under a voiceover track. Here are the basics:

First you will need to route your “Key Input signal” (the signal you want to be used to trigger the compressor), in this case your voice track, through a bus. Next, place the compressor on the music track to be “ducked” and set the compressor to side chain mode (also sometimes called key input mode). On the Digirack compressor clicking the button that has a key on it does this. Next you will need to let the compressor know what signal it is looking for to be triggered. Look for something that is labeled “key input signal” or similar. This will need to be set to the bus you assigned to your voice track. At this point all of the signal routing is complete and all that is needed is to tweak the settings to “duck” the music track based on the voice track. These settings will be very different depending on the tracks that you are using but I will typically use a fairly slow attack (100-150 ms) and very slow release or decay (600 ms – 1 s) to make this effect as unapparent as possible. I also use a fairly high ratio (6:1 or higher) to limit the music track when the voice track is playing and adjust the knee of the compressor (usually soft) to make the volume transitions as seamless as possible. Also, you will not need any gain compensation for this. The threshold will be dependent on the level of your recorded audio, but all of these parameters can be tweaked until it sounds right to your ears. Find what feels right to you and go with it.

Using a side chain compressor can save you a lot of time when mixing voice over on top of music and will certainly prevent the music from overpowering the voice. It can feel very confusing at first, but once you get it, my bet is you will find yourself using it on just about any project that has these two components.

Recording Tips: EQ

Continuing on through some basic recording techniques, this installment will focus on some basic equalizer (EQ) functions and how you can use this tool to enhance your recordings. As with the previous discussion on compressors I will focus primarily on the use of EQ for voice and discuss only very general aspects of usage. In it’s basic forms, the EQ has two different styles, graphic and parametric. The graphic EQ is the most basic and where this particular discussion with focus.

The audio that we record is comprised of many different frequencies coming through at different volumes. Equalization allows us to boost (turn up) or cut (turn down) those frequencies individually. Equalization done correctly can enhance your audio by giving it body (low end), warmth (mid range) and brightness (high end). Equalization can also correct minor recording issues such as plosives (sounds generated by bursts of air such as the “P” sound) and sibilance (hissing sounds such as the “S” sound). The EQ will not make the voice sound like someone else’s, but will enhance (good or bad) what is already there. To use a cooking analogy, the EQ is like salt; it can make good food taste great, but if you use too much of it, it will ruin it entirely.

Typically with a graphic EQ there are adjustable parameters for individual frequencies. Each one can be turned up or down to affect the audio. There is typically also a Q control, which determines how narrow or wide frequencies around the selected frequency will be affected. Think of the Q control as a gentle “U” shape with the affected frequency in the middle, the higher the Q control is set the narrower the “U” is, thus affecting fewer frequencies.

In most cases, the goal of the EQ is to be undetected by the listener, therefore cuts are much more friendly in facilitating this. Boosts can very quickly cause a voice to sound unnatural so boosting gently is preferred. You can take a little more liberty with the cuts without creating a too unnatural sound. Below are some general tips for voice equalization:

• Use of a High Pass Filter (HPF) reduces frequencies below the set frequency (this is sometimes called a shelf). This is very handy for use in vocals as most audible speech is below 125 Hz and applying an HPF somewhere below that can instantly remove plosives and rumble created from mic stand movement.
• Use a slightly lower Q setting (around 1.0) for a gentler, less apparent affect.
• Frequencies between 125 Hz – 1 kHz make up the bulk of the human voice; lower frequencies will affect the body of the sound, adding or subtracting richness while the higher frequencies will affect nasal qualities. Frequencies above 1.25 - 5 kHz affect the crispness of a vocal and allow it to cut through other sounds better such as a music bed. Frequencies above 5 kHz affect the brightness quality of a voice sometimes providing a more breathy sound. Too much boost in the higher frequencies will cause the voice to sound thin and brittle.

Not one recording is the same as the next, take time to work with each individual recording and address the uniqueness of the sounds. The EQ is a very easy to use tool and can have a very positive impact on your finished product if used wisely.

Recording Tips: Compressors

I have been asked many times to share my insights into creating better audio. It’s no secret that many more people today than ever before are doing some sort of recording on their own. The truth to this is that it is a great thing for those of us in the professional audio world (although some disagree). I receive innumerable projects across my desk from those that have recorded something on their own and want to improve it, add to it or work from it. Knowing how to use the basic tools to make better recordings allows these people to spend more time on the creative aspects of what they are recording and less time trying to figure out how to work all of the gadgets that are included in the recording software/hardware they purchased. Over the next few weeks, I will be writing about the usage of some of the most common audio tools used to achieve more professional recordings. I will be writing these as if the reader knows absolutely nothing regarding the tools, so if you are a little more seasoned, stay with me.

When working with voices, there are two pieces equipment that I almost always use, a compressor and an EQ. There are of course a whole host of other pieces that can be added to the audio chain, but these are the basics. For now, I will focus only on the compressor and save the EQ discussion for another day.

While probably the most commonly used audio tool, the compressor is a stumbling block for many people. For some reason, it just seems overly complicated when in actuality it is an extremely easy to understand tool. A compressor is essentially used to even out the volume of a sound source. Naturally, words that are spoken or sung using inflection and pitch will have areas that are louder and some that are much softer. A compressor, used correctly, will bring the space between the loud parts and the soft parts closer together by automatically turning down the loud parts and turning up the soft parts. In order to do this, you have to be able to tell it how to do its job. Before proceeding, I must first tell you that a compressor can do a lot of things and function in many different ways, today I’m only going to focus the most basic settings and controls. I will delve deeper into the other functions another day. The most basic user influenced parameters on a compressor are as follows:

• Threshold – This tells the compressor when to start working. The compressor will start increasing or reducing volume when the audio signal reaches the specified volume of the threshold. If the volume of the sound never reaches the threshold limits the user sets, it will not be affecting the audio at all.
• Ratio – The setting of the ratio determines how much the volume of the compressed areas will be affected. 1:1 being no affect at all, 20:1 being a very considerable amount of affect well past the point of acting as a limiter (also discussed further later). A ratio of 4:1 would mean that if the audio increases 4 dB above the threshold limits that were set the, volume increase would now only be 1 dB.
• Attack – This setting determines how quickly the compressor starts working after the set threshold has been reached. Attack times set to a very fast setting, say 1 millisecond, will almost instantly have compression applied once the threshold is reached, longer attack times will apply the compression more gradually. Remember, the goal when using these audio tools is not to be able to tell that they were applied, just to improve the sound, adjusting the attack time helps with this greatly.
• Decay – This tells the compressor how quickly to release the compression once the sound has crossed back under the threshold. Again, this is used to make this tool subtler. A fast decay time will release the compressor immediately while a slower decay time will gradually decrease the amount of compression.

Since every vocal recording is different, it is hard to give you a defined setting to use, but the compressor is actually very easy to manipulate and knowing what each control does should aid you in tweaking the effect until it sounds right. Your ears are the best tools you have; use them to compare your effected recording to others that sound good to you.