Digital Sound & Music: Concepts, Applications, & Science, Chapter 5, last updated 6/25/2013
method in that it discards information that cannot be retrieved, music lovers were quick to accept
the tradeoff between some loss of audio quality and the ability to store and exchange large
numbers of music files.
AAC (Advanced Audio Coding) is the successor of MP3 compression, using similar
perceptual encoding techniques but improving on these and supporting more channels, sampling
rates, and bit rates. AAC is available within both the MPEG-2 and MPEG-4 standards and is the
default audio compression method for iPhones, iPods, iPads, Android phones, and a variety of
video game consoles. Table 5.1 shows the filename extensions associated with MP3 and AAC
You may have read that MP3 offers a compression ratio of about 11 to 1. This number
gives you a general idea, but exact compression ratios vary. Table 5.3 shows ranges of bit rates
available for different layers within the MPEG standard. Typically, an MP3 compressor offers a
number of choices of bit rates, these varying with sampling rates, as shown in Figure 5.47 for
Adobe Audition’s MP3/mp3Pro encoder. The bit rate indicates how many bits of compressed
data is generated per second. A constant bit rate of 128 kb/s for mono audio sampled at 44100
Hz with 16 bits per sample gives a compression ratio of 5.5:1, as shown in the calculation below.
For stereo (32 bits per sample) at a sampling rate of 44100, compressed to 128 kb/s, the
compression ratio is approximately 11:1. This is where the commonly-cited compression ratio
comes from the fact that frequently, MP3 compression is used on CD quality stereo at a bit rate
of 128 kb/s, which most listeners seem to think gives acceptable quality. For a given
implementation of MP3 compression and a fixed sampling rate, bit depth, and number of
channels, a higher bit rate yields better quality in the compressed audio than a lower bit rate.
When fewer bits are produced per second of audio, more information must be thrown out.
Version Bit rates* Applications Channels Sampling rates
CD, DAT, ISDN, video games,
digital audio broadcasting,
music shared on the web,
portable music players
mono or
32, 44.1, and 48 kHz
(and 16, 22.05, and 24
kHz for MPEG-2 LSF,
low sampling frequency)
Layer I 32−448 kb/s
Layer II 32−384 kb/s
Layer III 32−320 kb/s
multichannels, multilingual
32, 44.1, and 48 kHz
Layer I 32−448 kb/s
Layer II 32−384 kb/s
Layer III 32−320 kb/s
AAC in
MPEG-2 and
Audio Coding)
8 −384 kb/s multichannels, music shared on
the web, portable music players,
cell phones
up to 48
32, 44.1, and 48 kHz
and other rates between 8
and 96 kHz
*Available bit rates vary with sampling rate.
Table 5.3 MPEG audio
Previous Page Next Page