Computers try to play noted music

Barry Vercoe dreams about conducting -- about the day he will flick his baton and a darkened concert hall will be flooded with music played not by some famous symphony orchestra but by a cluster of anonymous computers.

That day, the New Zealand-born composer readily admits, is still a long way off. But he insists it is only a matter of time -- and intensive research -- until computers can be made to respond, as musicians do, to the coaxings and commands of a conductor.

This project is the "next frontier" in electronic music for Barry Vercoe, a musician-composer who founded the Experimental Music Studio at the Massachusetts Institute of Technology.

Since 1973, Professor Vercoe has spent long hours trying to narrow the chasm between the intuitive, amorphous world of creating music and the hard-fact, scientific arena of computers. And for the average composer, Mr. Vercoe has succeeded in making the leap from arts to sciences a much less forbidding one.

Computer music is still a relatively young field, pioneered around 1950. And while a number of experimental-music studios have cropped up, only a handful are regarded as major centers -- MIT, the Institute for Research and Coordination Acoustics/Music in Paris, Princeton University, Stanford University, and, in the not too-distant future, the University of California at San Diego.

What sets the MIT studio apart is an innovation that allows a composer to communicate with a computer in his own language -- playing notes on a keyboard that looks like a piano.

At virtually every other experimental-music studio, composers must type scientific terms on the computer, which are translated into punch cards and teletypes which, in turn, produce the music. Mr. Vercoe's students compose with the tool most familiar to them, a piano keyboard. The computer is hooked up to the keyboard. It then translates each note that is played into its own complex language and records the note on its screen in a musical score, complete with clef lines.

For the average musician, who is not likely to be a computer scientist, such a luxury means that an elusive melody or half-finished passage can be quickly captured with the plunk of a key and not lost in the painstaking process of translating a musical note into computer language.

In the three-room suite where MIT's small band of budding composers works, the air is often flooded with sound rippling from stereo speakers in each corner of the composing room.

In that small, box-size room, serious composers -- not computer tinkerers who want to fiddle with sound -- work with computers to copy the sound of familiar instruments like violins and horns or, better yet, to conjure up new "instruments" or music of their own.

Mr. Vercoe sees computers not as a replacement for traditional instruments but as another medium for composing.

But making music with computers has benefits Mozart never dreamed of. For one, he says, intricate passages of quicksilver notes that would be virtually impossible for a human performer to play are only a matter of proper programming for a computer.

A composer in an electronic-music studio has a luxury more traditional colleagues do not have: being able to throw a switch and, with very little delay , hear the music just composed played back by the computer.

MIT is one of the few labs that rely on "digital synthesis" -- a finely honed , precise system that sunthesizes sound as numbers. The more traditional "analog synthesis," the process used in Moog synthesizers, translates music as continuous waves or voltages.

"The limitation to analog synthesis is a restriction to the degree one can get into fine scalpel work," Professor Vercoe explains. "The tools are not fine enough to get in there and really dissect and change a sound. Composers are forced to make changes on a larger scale.

"With digital, it's done not with waves, but with a series of little numbers that go up and down, just the way a sound wave does. We can change, add, and subtract numbers to restructure sound."

Still, the world of computer music is not one of total harmony. Many problems remain to be solved -- some of which are minor, such as how to stage a concert that has no live performer.

"For the average audience, it's a problem. what do you do when it's over? Just sit there, or applaud the computer?" says Mr. Vercoe, whose own computer-music scores have been heard at Carnegie Hall and the Kennedy Center.

The solution, he suggests, may be to set up computer concerts like an art show, where listeners can wander from computer to computer, savoring each piece for as litte or as long as they like.

There are, however, more prickly questions. While Mr. Vercoe may have succeeded in making music students more at home in a computer lab, electronic music still hovers in a no man's land between the arts and sciences.

Only 15 years ago, he says, a serious composer who even considered working with synthesizers was putting his musical reputation on the line. And though today computer composers are no longer the new kids on the block (the MIT studio has received funding from both the National Endowment for the Arts and the National Science Foundation), they are still regarded somewhat suspiciously by music and science purists alike.

And even with all the promise of computer music, there remain limitations as well. Theoretically, Mr. Vercoe says, digital synthesis means that musically speaking, anything is possible. The only hang-up? Man-to-computer communications.

Although the computer is capable of creating any kind of music, it can do only what the person sitting at its keyboard tells it to do. The result is an at-times lengthy, trial-by-error process during which the composer roughs out an idea and the computer plays it back, until at long last the composer has the sound he wants.

"You just can't tell the computer, "Play me a nice note,'" Mr. Vercoe explains. "The computer doesn't know how. It reminds me of an old Jimmy Durante line. He'd say, 'Stop the orchestra. Trumpet, play that note. That's a good note.'

"That's the whole problem. What is a good note?" he continues. "We don't even know how to describe a good note to each other, let alone to computer. that's a big problem, and it will be a big problem forever.

"But," he adds, "that's always been the fun of composing."