Basically, compressing a vocal line is either quieting the loud parts, amplifying the quiet parts, or both (usually both). This makes the overall difference in volume less, so we can hear the entire line over the music (not just the loudest parts).
Most vocals need compression to be heard over music of even medium volume or intensity (and you might even want to compress a capella vocals too). Compression can be easily done with a software compressor (because if you’re reading this, you don’t have a hardware compressor). Where do you download one? You probably have one if you’ve done anything with VOCALOID in the past. Everyone’s favorite free audio-editing tool Audacity has a nice one built right in! Now, of course, Audacity’s compressor is nothing to write home about, but it definitely beats uncompressed vocals. I prefer FL Studio’s Fruity Compressor for my stuff, but hey, I’m no pro either. Whichever program you use for audio editing, I can almost guarantee it has a compressor built in; look it up online if you can’t find yours.
To work any given compressor, you need to know a few things and what to do about them.
- Gain is the volume you add to the track. It’s measured in dB (decibels). 0.0 is the original volume, with positive numbers indicating an increase in volume and negative numbers indicating a decrease.
- Threshold (also measured in dB) is the volume at which you start applying the compression. A lower threshold means you’re affecting softer parts of the track, and therefore more of it.
- The Ratio is how much you are compressing the track’s audio. The original audio track has a 1:1 ratio of volume (soft parts are soft, loud parts are loud). By increasing the ratio, such as to 4:1, you take all parts of the track above the Threshold volume and flatten them out so that each 4 dB increase is lowered to a 1 dB increase. Likewise a 4 dB decrease is lowered to a 1 dB decrease. (this makes the soft parts louder, and the loud parts softer).
- Attack (measured in milliseconds) is the time it takes for the compressor to apply the Ratio to the track. A shorter attack means a more rigid change in the volume. If you don’t think you want to mess with this setting, then leave it at the default.
- Release (or Decay, also measured in milliseconds), similar to Attack, is the time it takes for the compressor to stop applying the Ratio to the track.
Don’t worry if you find all this too confusing at first, because I’m going to give you the easiest answer. Vocals are often compressed quite hard, meaning their soft parts and loud parts are not that different in volume (but this doesn’t, for example, change the sound of a whisper into a scream or vice versa). In most cases, you’ll want to compress vocals with a high Ratio and a low Threshold. A good place to start would be:
Ratio: Between 6:1 and 9:1. ( I like to compress VOCALOID vocals extra-hard because their consonants tend to get swallowed.)
Threshold: Around -40 dB. (Any higher and you’ll start to lose your quietest sounds, and we don’t have a problem with background noise because VOCALOID doesn’t have any!)
Gain: Depends on your background music, so play with it. Audacity’s Compressor has a neat “Make up gain for 0dB after compressing” check-box that you can leave checked instead.
After you compress your vocals, check to see if you can hear every part when you place it over your background music. If not, try raising the gain a bit more (In Audacity, that’s the little minus-to-plus slider at the front of the track). If it still doesn’t work you’ll have to play with the settings a bit, or maybe you just need to continue the mixing process by moving to the next section of this guide.
Continued on to the Next Page »
Onward to Equalization!