The following is a paper Tim wrote for a computer science class as an undergrad at Wartburg College in Waverly, Iowa. Many of the details are out of date (it was written in early 1992, and many of the sources are from the 1980's), but the basics still hold true, and it contains some historical information that many of you might find interesting.

Computers and Music

By Timothy S. Fischer
(1992)

There is no escaping electronic music. Whether it be television commercials, popular music, movies, or kid's cartoons, it's safe to say that most contemporary music has some sort of electronic instrumentation. Pick out a popular song at random, and there's a pretty good chance that a computer was involved in its creation (Yelton 11).

Synthesizers, the instruments which create electronic music, have been around for some time. The first attempt at harnessing the computer as a tool for synthesizing sounds was in the mid 1950's. Bell Laboratories of New Jersey became interested in the possibilities of transmitting telephone conversations in digital form. This process involved converting the conversations from analog to digital form by a process called sampling. Then at the other end, the digital information was converted back into analog form. This system needed only to transmit a very small portion of the audio spectrum, but the team realized that it might be possible to use similar technology to create music electronically. In 1957-58, this research resulted in MUSIC I and MUSIC II, two very simple computerized music generation programs (Manning 217).

The next significant impact in electronic music was the work of Robert Moog and others in the mid-1960s. Moog fashioned musical instruments from electronic lab equipment, and some of his instruments appeared on popular records by Walter Carlos, the Beatles, the Monkees, and others. Rock music quickly incorporated synthesizers, and the analog synth sound of the 1970s laid the foundation for many hit records. In the last few years, new techniques in digital synthesis have made many new sounds available to anyone who wants to experiment with electronic music (Yelton 13-14).

Analog synthesizers operate through electronic oscillators and filters that generate simple combinations of pure frequencies with standard waveforms, such as sine or square. These sounds may be interesting in their own right, but only superficially resemble acoustic instruments. Digital instruments, however, use a sampling process, which uses digital recordings of real instruments. When a musician presses a key, the machine calls the digitized note from memory, and sends it through a loudspeaker. This process in theory can mimic any acoustic instrument, except sampling can use a lot of memory (Brody 27).

For example, to accurately mimic the sound of a piano would require 25 billion bits, filling 100,000 memory chips. To avoid this cost, sampling instruments often cut corners. Many keyboards store only the initial "attack" portion of the note, and a short portion of its final decay. Another shortcut is to record only a fraction of an instruments notes, say 20 or 30 of the 88 notes on a piano, and then electronically raise or lower their pitch to produce the remaining notes. Unfortunately, both of these methods result in distorting the sound, which makes it differ from the acoustic method. One recent method of condensing sampled sound was used in the Kurzweil 250 keyboard. This keyboard used a set of rules that describe how the instrument sounds without actually "spelling it out". This method resulted in the most realistic sounding keyboard yet (Brody 28-29).

Although the most common electronic instrument is the keyboard synthesizer, some new and unusual instruments are appearing in the last few years. Along with electronic wind instruments such as the Yamaha WX-11 Wind Controller used in the last Wartburg Artist Series, and guitars such as the SynthAxe which can control traditional synthesizers. The SynthAxe has two sets of strings: a short set of strings to trigger notes in a synthesizer, and a long set to vary the notes' pitches. This instrument allows for some unique musical effects, such as "strumming a brass section" ("Electronics" 38).

There are also some other unusual instruments. Philippe Guerre designed an instrument he calls the Laserharp. Consisting of a laser and a rotating mirror, along with photoelectric sensors which detects whether or not the laser beam has been interrupted and where in the path of this beam the interruption has taken place. The beams of light can be "plucked" just like a harp, and the pitch changes as the musicians had moves up and down the beams. Guerre also has a contract with the French space organization Aerospatial to launch the Laserharp software into space in a satellite. The satellite would then use the light from stars to play melodies (Connor 37).

One of the most significant developments in the history of electronic music came in 1983, when the country's major electronic-instrument manufacturers announced a communication standard called Musical Instrument Digital Interface, commonly referred to as MIDI. Before MIDI, it was not uncommon to see a keyboardist playing several instruments at once. Since various keyboards have their own strengths and weaknesses, they weren't able to choose one over another. Keyboards couldn't be interfaced, because there were no standards. However, MIDI changed all this. MIDI avoided compatibility problems by largely ignoring how an instrument produces sound. Physically, MIDI consists of three wires with a five-pin DIN connector at each end, which plugs into the instruments. A MIDI instrument does not directly control another. Rather, the first sends information to the second describing when to make a sound, what pitch and instrument the sound should be, how long the sound should persist, and other important information. The controlled instrument simply does what it is told, following the lead of the first. For example, when a key is pressed on the controlling synthesizer, three bytes of information are generated. The first indicates a key has been depressed, the second indicates which key was depressed, and the third indicates the velocity of the depressed key, or a dummy value if the keyboard doesn't support velocity sensing. Dissimilar instruments can be interfaced, since the simpler instrument can ignore any MIDI information it cannot use. For example, many higher-priced instruments use the velocity of the key pressed to indicate how loud to produce the note. If the controlling keyboard produces his information, and the controlled keyboard cannot use it, it is simply ignored (Jordahl 79; Hall 27; "Electronics" 38).

Obviously, MIDI was designed for the performing musician, but the same interface allows keyboards to be connected with personal computers, leading to many different applications. Since MIDI was designed with professional musicians, it's not surprising that there have been some major benefits to them from this marriage of computers and musical instruments. One of the largest benefits of combining computers with keyboards comes with music writing. With music notation software, such as Finale on the Macintosh, or Score on the IBM PC, music can be played on the keyboard or other MIDI instrument, and immediately be displayed on the computer screen. Musical pitches can also be entered with a mouse or the computer keyboard. Mistakes can be corrected, and other musical symbols, such as dynamic markings, tempo changes, and even lyrics, can be entered to make a complete score. When finished, the score can be printed out. If a laser or other high-quality printer is used, the music approaches published quality. The advantages of using music notation software parallel the advantages of using a word processor rather than a typewriter (Mahin "Choosing..." 17-23). My brother, Michael Fischer, is a professional musician who is just getting started in music composing. After writing several compositions by hand, he has recently switched to a program called Music Prose. Fischer said that initially you don't save that much time using the computer compared to conventional hand-notation means, but this time is made up if the piece needs to be edited. For example, when he sent a piece in to be published, they recommended some changes be made. Since he had written this piece out by hand, he had to rewrite the entire score to make these changes. Had he have written this piece with Music Prose, he would have simply had to make the needed changes and reprint.

Other applications for professional musicians include software known as librarians. These programs organize and store synthesizer sounds, known as patches. Synthesizers have a small amount of memory, which can hold a finite number of patches, usually in a bank of 32, 64, 100, or 128 sounds. However, if you have a patch librarian for the computer, you can store thousands of synth patches on a single floppy disk. This information can be downloaded, via MIDI, from the synthesizer to the computer, and they can be reloaded to the synthesizer when needed (Yelton 149-150). Some librarians, also voice editing software to help create new sounds. Since changing these settings on the synthesizers themselves can sometimes be hard to use, voice editing software, such as the Opcode Editor/Librarian, allow these settings to be made conveniently on screen with a mouse (Mahin "Software..." 49).

In addition to the benefits to professional musicians, there are also many benefits for those who teach music and their students. One of the most popular MIDI applications for the classroom is known as sequencing. Using sequencer software, the computer acts like a tape recorder. However, instead of recording the sounds, the sequencer records the MIDI information which tells which notes were used, how they were used, etc. This data can be stored to disk or edited in a variety of ways. For example, tempo can be changed without changing the pitch. Also, passages can instantly be transposed to another key. Sequencers can also record and play back different parts of a composition in a multi-track manner. Therefore students can record a basic chord pattern, then listen to it while adding a melody line. Then these two parts can be played while a bass part is added, and so on. When the process is completed, the computer can then print out a complete score. (Jordahl 79-80; Blackman 29). Sequencers are also helpful for professional musicians who are composers. If the computer is connected to a synthesizer which can make the sounds of the instruments the composer is calling for in the composition, the composer can hear what the piece actually sounds like before it is fully completed (Ehle 45).

Another use for computers and keyboards for students is in individualized music instruction. For example, students can use the synthesizer to play a musical passage shown on the computer screen. The computer "reads" the input data, and watches for errors in notes, rhythm, or tempo. Incorrect input is shown in musical notation next to the original passage, while both are played back for the student to compare. Similar programs are becoming available for other instruments besides keyboards. At least one software series currently on the market is designed especially for singing instruction. Students sing passages into a microphone, and the computer then indicates if the notes sung are flat, sharp, or on pitch, while a synthesizer plays notes for comparison (Jordal 80; Hall 28).

Computers are also helping people who are physically disabled write and play music. With a MIDI keyboard and sequencer, people who are moderately disabled-- that is they can coordinate one hand or even just a finger, can compose and play music. At the present time, the severely handicapped cannot take advantage of MIDI on their own, but need someone to help. Research in this area is important, because if disabled children cannot communicate, then no one can understand what they think. Before computers, severely disabled children could only communicate by "eye-pointing", in which the "talked" by focusing their eyes on different symbols to make simple sentences. Today, word processing programs for the disabled allow more sophisticated communication. A program might scan a cursor over a grid of letters, words, numbers, etc. A switch that is triggered by the disabled person stops the cursor and sends whatever is beside it to the top of the screen to build up a sentence. Drake and Grant would like to see a similar system developed for music (Drake and Grant 37-40).

Another system, known as Biomuse, could also eventually help disabled persons. The Biomuse system consists of dime-sized electrodes which are hidden underneath a headbands and armbands placed on the musician. These electrodes pick up signals from the musician's brain and, eye, arm, and hand muscles. A small box of newly designed circuitry would gather, filter, and process the bioelectric signals so they could control a music synthesizer through the MIDI interface. What this all means is that a musician could walk on stage, and begin playing: with every gesture of her hand, every flick of a finger, notes would emanate from the banks of speakers, and when she closes her eyes, an unseen piano would accompany the melody (Amato 202-3). In addition, this system is ideally suited for someone who has very little motor control. Since the electrodes actually work on brain waves, totally disabled persons might be able to make music without moving a muscle. The Biomuse technology may also help disabled people get jobs by enabling them to control computers without needing fine motor control. The inventors of Biomuse are looking into the possibility of using the signals created by the eye muscles to move the cursor around the screen like an "eye-controlled mouse".

As you can see, the potential of combining music and the computer is only beginning. The MIDI interface was only established in 1983, but as can be seen by its mention in almost every aspect described in this paper, it is indispensable, and new applications are being found constantly. Although I personally don't believe traditional acoustic instruments will ever become obsolete, I feel we will see electronic and computer music and its applications become more and more prevalent.

Works Cited

Amato, Ivan "Muscle Melodies and Brain Refrains: Turning Bioelectric Signals into Music" Science News V135 (April 1, 1989): 202-203.

Blackman, Jaimie M. "The MIDI Potential" Music Educator's Journal V73.4 (1986): 29.

Brody, Herb. "Kurzweils Keyboard" High Technology V5.2 (1985): 27-32.

Campbell, Phillip. "The music of digital computers" Nature 324 (1986): 523-28.

Connor, Steve. "The light way to play with the stars" New Scientist 112.1530 (1986): 37.

Drake, Adle, and Grant, Jim. "Music gives disability a byte" New Scientist V133.1544 (1987): pp. 37-39.

Ehle, Robert C. "Musicians and Computers" The American Music Teacher V35.5 (1986): 30-31, 45.

"Electronics Develops Charms to Sooth" New Scientist V113.1544(1987).

Hall, Vann. "Conquering the MIDI Muddle" Music Educator's Journal V73.4 (1986): 27-9.

Interactive Music Company, The Book of MIDI Computer Software (Hypercard Stack). Distributed by Opcode Systems, Inc., for the Apple Macintosh.

Jordahl, Gregory. "Teaching Music in the Age of MIDI" Classroom Computer Learning V9.2 (1988): 79-83.

Mahin, Bruce P. "Choosing Music Notation Software" Clavier V28.6 (1989): 17-23.

Mahin, Bruce P. "Software Review: Opcode Editor/Librarians" Instrumentalist V45.9 (Apr. 1991): 49.

Manning, Peter. Electronic and Computer Music Oxford: Oxford University Press, 1985.

Yelton, Geary. Music and the Macintosh Atlanta: MIDI America, Inc. 1989.