Understanding Voice Encoding

When you speak into the mouthpiece of a telephone headset, your audio input is initially
sent as an analog transmission over the telephone wiring. When the analog transmission
reaches the entry point of the PSTN, it is digitized or converted into digital format – a series
of zeros and ones. Once is has been digitized, the encoded voice transmission is transported
across the PSTN network to the far edge, where it is converted back again to analog.
The method for converting audio into digital has been standardized. The name of this standard
is G.711, and it uses an encoding technique called pulse code modulation (PCM). But,
within the G.711 standard, there are two varieties:
G.711u: Also known as µ-law encoding (the Greek letter “mu”), this is used primarily in
North America.
G.711a: Also known as a-law encoding, this is used primarily outside North America.
G.711 converts analog audio input into digital output at an output rate of 64000 bits per
second, which is commonly referred to as 64 kilobits per second (kbps). A single G.711
voice channel is referred to as “digital signal, level 0,” or DS0. The fact that a DS0 takes up
64 kbps has been used in building links of the PSTN. Thus building a phone network link
with a capacity for 24 voice channels would take 24 x 64 kbps = 1.544 megabits per second
(Mbps). A link with this capacity is known as a “trunk level 1” or T1 link.