Audio normalization is the process of increasing (or decreasing) the amplitude of an entire audio signal so that the resulting peak amplitude matches a desired target. Dynamic range and signal to noise ratio remain the same as the original mix.
Mastering has become a critical step in producing a CD or DVD. Normalizing is commonly used when mixes are assembled. Often, different tracks have been recorded and mixed differently. Each sounds fine in isolation, but when put together there can be large differences in volume levels. Normalizing allows you to set the volume of tracks on your CD to a standard level.
How Does Normalization Work?Edit
Normalization software will generally scan the track to determine the highest peak levels. It then calculates the ratio of the highest levels to the target level. Then the software will proceed to modify the samples up or down by the calculated amount.
If, for example, one track has a maximum peak level of 90% of the target then all samples will be raised by 5% (assuming the target is 95% of the maximum possible range).
If another track has a maximum peak level of 105% of the target range then all samples will be lowered by 5%. Notice that niether case changes the dynamic range or signal to noise ratio.
Normalization does very little when there is a maximum level already close to the target. If you require an overall volume change in this case you will need to use compression instead.