Audio

Knowledge Base | AudioMixing >

This article is about Audio theory; see also audio mixing and Audio CDs. Audio is not to be confused with sound.

A professional microphone

Audio is the transcription of acoustical energy (sound), most commonly into electrical impulses which can be conducted, stored, duplicated or manipulated by electronic circuitry. In rare occasions, audio may also be stored by mechanical means, such as phonograph records.

Confusion between sound and audio is a cause of much ambiguity, in normal speech the terms tend to be used interchangeably; strictly, though, audio refers only to the transcription of acoustical energy and sound refers only to the acoustical energy itself.

With the advent of digital computers, audio can be manipulated extensively and duplicated endlessly without generation decay. Computer technologies have caused a revolution in audio technology and practice, and have provoked major disruption of consumer companies engaged in the sale of audio media, particularly the music industry.

The transduction of sound into audio is known as "recording" and the transduction of audio into sound is known as "playback". Various properties of the audio waveform can be manipulated before playback or after recording and different audio signals are sometimes mixed

Changes to "equalization" (also known as "EQ") alter the frequencies of the wave, which cause a difference in how it sounds. Changes in amplitude cause the wave to be percieved as "louder" or "softer". These changes can be made with analogue electricals, such as potentiometers, or by digital means, with electronic circuitry and software. Both means have their advantages and disadvantages

Simplified diagram of how Pulseaudio works, in theory. Audio lib-
raries and the multi-media subsystem have been omitted for clarity.Audio input/output is produced on computers by a printed circuit board called a "sound card" (not to be confused with a "sound board" which is professional mixing hardware). The sound card manages the exchange of electronic impulses between physical inputs and outputs such as speakers and microphones, and interfaces with the operating system's multimedia subsystem.

On UNIX systems, popular multimedia subsystems include ALSA "Advanced Linux Sound Architecture" and OSS "Open Sound System". Both of these are "kernel modules" which are embedded directly into the heart of the operating system and enable the kernel to know how to interact with a sound card.

Due to technical limitations of the system design it is not possible for more than one program to access hardware directly. Ordinarily that makes it inconvenient to output sound from two or more programs at once. One solution to this problem lies in a sound-server such as Pulseaudio, the most popular.

The UNIX sound stack

Sound servers run as a daemon in the background and access the multimedia subsystem, mediating and multiplexing input/output from user programs and libraries. This works perfectly in theory but in practice a number of problems remain:

• Sound servers are an additional source of latency.
• Sound servers do not work perfectly.
• Many programs exist that access sound hardware directly.
• Operating system vendors have implemented sound servers very poorly.
• A multiplicity of sound servers exist, with incompatible interfaces, necessitating unduly complex workarounds and forcing developers to develop for multiple servers.
• Some types of software are inherently unsuitable for use with sound-servers.

K.B. Index | AudioMixing >