275
Appendix C: Glossary
A
ACM (Audio Compression Manager) A Microsoft technology that enables Windows
applications to compress and decompress files in a variety of formats, such as DSP Group
Tr ueSpeech and GSM 6.10. Some ACM formats install with Windows, while others install
with software applications.
ADPCM (Adaptive Differential Pulse Code Modulation) An audio compression scheme
that compresses sound files from 16 bits to 4 bits, yielding a 4:1 compression ratio. There
are many varieties of ADPCM, such as the IMA (Interactive Multimedia Association) DVI
standard, and versions from Microsoft, Dialogic, and others.
ActiveMovie See “DirectX” on page 279.
Adapter A cable, plug, or jack that enables you to connect two audio or video devices
together.
ADAT A digital 8-track tape deck manufactured by Alesis Corporation that is very popular
in recording studios.
Aliasing Noise that occurs when a high frequency sound exceeds the Nyquist Frequency
for a given sample rate. (See “Nyquist Frequency” on page 284.) Most analog-to-digital
converters prevent aliasing by filtering out sounds above the Nyquist Frequency.
Amplitude Amplitude represents the loudness of an audio signal. A waveform’s amplitude
is measured by its distance from the center line, which represents an amplitude of 0. There
are different standards for measuring amplitude, but the decibel (dB) is the most
common. (See “Decibel (dB)” on page 279.)
Analog recording Tr aditional audio recording with devices such as magnetic tape
machines and vinyl records. Analog audio recording consists of a continuous curve, as
opposed to digital recording, which consists of discrete samples.
ASCII text data You can represent audio data in this standard text format (.txt), with each
sample separated by a carriage return, and channels separated by a tab character. Before
the audio data, you can add a header with a format of
Keyword:Value, with the keywords
Samples, BitsPerSample, Channels, SampleRate, and Normalized. (The values for
Normalized are True or False.) If no header exists, the data is assumed to be 16-bit signed
decimal integers.
ug.book Page 275 Tuesday, March 16, 2004 1:29 PM
