ABSTRACT
Time domain sound data inside a digital audio process is a set of numerical values spaced at regular intervals. Each value represents the amplitude of the signal at a particular point in time, as described in Chapter 2. At a sample rate fS , each sample lasts
1Ef S seconds. It
is possible to convert between lengths of time and numbers of samples as follows:
length of time (in seconds) equivalent to 1 sample G 1
length of time equivalent to i samples G i
number of samples equivalent to T seconds G T F fS
It is useful to be able to calculate the total data size for a particular length of time, using an equation like this:
total data (bytes) G T F fS F C F B
where T G time in seconds
fS G sample rate
C G number of channels
B G bytes to store 1 sample
The trend in process implementations over time is towards higher sample rates, larger numbers of channels, and higher precision data representations (and thus the size in bytes of each sample). The variables in the equation above not only affect the amount of storage required (such as for an audio recorder), but also the computational power required to process this data in real-time. For example, increasing the sample rate from 48kHz to 96kHz requires a doubling of the amount of data processed per second.
