Howdy.
Compression of audio, be it vocals or a complete stereo mix is this: It basically reduces the 'dynamic range' of it, so that instead of quiet bits and loud bits, it evens them out so that everything is at a similar level.
Effectively, its process is that it squashes the loud bits down to a similar level to the quiet bits, then pulls the whole lot up. The end result is that the perceived volume will be louder, and at a constant level.
The basic settings you'll see on a compressor will be the following:
Theshold - the level at which the compressor will start to compress. For example, if you set a threshold of -10dB, when your level tries to go over -10dB, the compressor will reduce the range according to the ratio.
Ratio - this is effectively how much it compresses it... You'll see ratios quoted as 2:1, 2.5:1, 3:1, 4:1 etc. etc. For example, with a 2:1 ratio, for every 2dB the level tries to go over the threshold level, it'll only go one. At 3:1 ratio, for every 3dB it tries to go over, it'll only go one, and so on and so on. So, the higher the ratio, the 'more compressed' it would be.
Attack / release - how fast or slow the compressor starts working and stops working. A slower attack will leave some spikeyness, a faster one will keep it more even. With a slower release, the compressor works for longer, and so level is kept down for longer. Generally these setting are measured in miliseconds.
Gain - as the compressor squashes the level downwards, so that the loud bits are as quiet as the quiet bits, the whole thing needs to turned up, hence gain control.
You'll find other settings on various compressors, but the above are the most important...
