INART 55

History of Electroacoustic Music

Audio and Personal Computers


Digital Audio Workstations (DAWs)
Shortly after MIDI was introduced, personal computer audio editing software/hardware systems began to appear in 1985.


The first PC audio editors were introduced by the Digidesign company in 1985, with a program called Sound Designer that ran on the Macintosh computer. It required a large hard drive and extra hardware to do AD-DA conversion. It was an expensive system, but it became a big seller as studios quickly began adopting digital recording as an alternative to tape. Not only was the graphic interface helpful for editing, but edits were non-destructive: unlike tape, an editing operation could be undone with no loss in quality.

Digital Concrète
Pierre Schaeffer's research center, GRM, also marketed its first software for audio manipulation in 1985. Thus, musique concrète became mainstream, as more and more people began to use computers to edit audio.

Csound
In 1984 MIT's Center for Experimental Music became the Media Lab. Founding member Barry Vercoe focused on Music Cognition and Machine Understanding. Having gone to IRCAM and seen the 4X computer, he determined that the computer should also be programmed to "think" about the music it was creating, and thus make intelligent decisions about how it was responding to the music of performers (recall Max Mathews' idea of the intelligent machine). In 1985 Vercoe created a version of Mathews' Music V software written in the C programming language. He called this new incarnation Csound. In keeping with Mathews' spirit of openness, Csound remains available for free, downloadable from a number of Web sites, with versions available for virtually every computer hardware platform. The source code is also available, allowing programmers to customize it by creating their own modules for it.

Pentium and PowerPC
In 1993 the Pentium microprocessor was produced. This contained 3,100,000 transistors, and ran at unprecedenteed speeds. The Pentium processor was created by Intel, while a counterpart, the PowerPC, was created by Motorola. By 1995, these processors were standard for PCs, allowing any PC to function as a digital audio editing workstation without special hardware for conversion. This represented a quantum leap forward for the personal computer, which now was virtually the equivalent to UNIX systems in speed and power. The age of desktop multi-media was born.

By the end of the 1990s, most synthesizer manufacturers were no longer producing synthesizers. Rather, they produced stage pianos, with some timbral control, but nothing near the capabilities of a full synthesizer. This was all most keyboardists wanted. At the same time, software synthesizers (or "soft synths") have become increasingly popular. In addition to Csound, there are numerous soft synths on the market for those who wish to work with sound synthesis.

In 2001, Apple released OSX, the first personal computer operating system based on UNIX. Personal computers were now not only equivalent in power to UNIX computers, they were the same thing. UNIX was no longer the more exclusive and elite system of the research community. A version of UNIX for Intel-based computers, LINUX, had been available for some time, but had been incompatible with many aspects of the Microsoft operating system that dominated the Intel platform. These difficulties were gradually addressed, making UNIX a powerful player in both dominant personal computer operating systems.

Karaoke and the Future
At the turn of the century, another important area of development was funded by Karoake manufacturers. Barry Vercoe's Csound software was the basis of a multi-processor system that was implemented in professional karaoke systems produced in Japan. This version allowed continual tracking of a singer's pitch and tempo, plus a full array of audio effects such as a phase vocoder, which allowed prerecorded backing vocals to follow the karaoke vocalist. Csound was also the basis for a new standard known as MPEG 4 Structured Audio. This functions in a manner similar to the Postscript printing language, by which a series of commands are sent to a printer, which decodes them to realize a printed page. With the Structured Audio protocol, efficient commands to realize audio can be sent over the Internet, so that audio at one location can easily be encoded, transmitted, and decoded at the receiving end. Different decoders are employed for different types of material (drums, vocals, synthesizers, etc.) to ensure maximum fidelity. This will allow users to receive a number of discrete tracks and to mix them as they please, whether the intention is to produce music for karaoke or any other purpose.