Using audio compression the right way for professional sound

Video has a dynamic range, meaning the difference between the brightest brights and the darkest darks. Within that dynamic range, you can distinguish image detail. Outside that dynamic range, you’re seeing only pure white or pure black, and no detail at all. We call this video clipping.

Sound also has a dynamic range, the difference between the loudest louds and the softest softs. Any sound outside of that volume range becomes either inaudible (too soft) or distorted (too loud -- we call this audio clipping). Either way, you lose any detail in the sound.

A compressor limits that dynamic range, in order to prevent distortion or to boost detail in the lowest lows or the highest highs.

Compressing sound is a lot like adjusting exposure in video to keep it within a certain dynamic range and preserve detail in the lowest lows and highest highs.

In one respect, a compressor works similarly to an audio limiter in that it limits the top volume level, keeping volume at or under a specified level to avoid clipping. But a compressor gives you more control than a limiter alone. As a compressor limits the top volume level, it can simultaneously bring up the lowest volume levels. This process of limiting the loudest levels while raising the lowest levels thus compresses the audio dynamic range. The entire range of sound volume now sits within modified minimum and maximum levels designed to bring out all audio while avoiding clipping.

Threshold: the volume level at which compression engages. The plug-in applies compression to any sound which exceeds the threshold, but leaves the rest of the sound alone. A tight compressor doesn’t let any sound get louder than this point; a looser compressor allows some flexibility. Threshold works together with the ratio, so adjust them together for best results.

Ratio: the amount of top compression applied, or how far down the compressor pushes the loudest sounds. A ratio of 1:1 basically means no compression. A ratio of 2:1 means the unaffected sound must rise 2 dB past the threshold in order for it to rise 1 dB after compression is applied. A ratio of 4:1 means it must rise 4 dB past the threshold in order to rise 1 dB after compression. The higher the ratio, the more compression, and the more it limits the top volume. Generally, go for a softer touch, and then apply more as needed. A light compression ratio like 2:1 often works best for rich sounds like human voices. A heavier ratio of 4:1 or 6:1 may be better for bass-heavy sounds.

Attack: how quickly the compressor engages and lowers the volume. You can set this to near-instant lowering, or you can let the compressor lower the volume more gradually, just like if you turn a volume knob down really quickly or more slowly. Faster attacks work best on sounds or voices with sudden peaks, while slower attacks work better on smoother sounds, like string instruments.

Release: how quickly the compressor lets the volume return to normal once the volume falls back beneath the threshold. Like attack, you can set this to near-instant or gradual, just like turning a volume knob back up quickly or slowly. Slower release times can sound more natural.

Input Gain: lets you adjust the overall level of volume of your sound before compression.

Output Gain: lets you adjust the overall level of volume of your sound after compression.

For final mastering of your video, do all of your other audio work first, including whatever individual compression you may need or want, then do a final compression pass on a master audio bus.

Often, you want to compress according to how your target audience will hear your video. Will most listen through larger speakers like headphones, bookshelf speakers, home theater systems, or larger computer systems? Will they listen through earbuds or laptop speakers? Each kind of speaker has its own dynamic range, and good use of compression keeps your audio within that dynamic range.

If you compress with earbuds in mind, you generally compress for tighter dynamic range, keeping the highs from slicing and the lows from distorting or disappearing altogether. Laptop speakers tend to have better dynamic range, but certainly not the dynamic range of headphones, bigger computer speakers, bookshelf speakers, or a home theater system.

But even if you don't intend for your viewers to listen on earbuds, it’s still a good idea to hear how your audio sounds on them and try to eliminate any problems, because some of your audience may well listen that way.

Since just like with video, VEGAS Pro edits audio non-destructively, your compression work will never affect your original audio files. That means you can fix mistakes easily. Just reset the parameters or remove the effect completely. You can’t destroy your original audio files, so experiment. Try anything you like! You may find an incredible sound that no one’s ever found before. At the very least, you'll soon learn to use compression carefully, strategically, and tastefully to improve the quality of all of your audio productions.