GOES DIGITAL
by Irv Teibel
Irv Teibel, creator of the Environments series of psychoacoustically based audio recordings, owns an Apple II and two used Cadillacs.
The ultimate ocean on vinyl.
A good deal of the progress of civilization over the centuries has been a function of gaining control over natural processes so they can be used when we need them. Some examples are windmills, the internal combustion engine, atomic power and Donkey Kong.
In 1968, with the help of a computer, I made a modest contribution to this august confluence of imaginative derring-do by putting the true sound of the ocean on a phonograph record. At the time I often played chess with a genius friend whose profession of the moment was psychoacoustics, an arcane science seemingly focused on band-aid fixes of airport noise and improving intelligibility of telephonic transmissions. One night he happened to mention some reading he'd done on Hermann Ludwig Ferdinand von Helmholtz, a nineteenth-century German scientist who felt that natural sounds-the ocean, wind, rain and other mundane sonic occurrences-could have great psychological benefits if only some means of accurate reproduction could be found.
It happened that I had just returned from a wintry sojourn at Brighton Beach, where I had made a brief recording of ocean waves for the sound track of a friend's underground film, and this casual mention of Helmholtz' musings triggered a "what-if" that was to have a profound effect on the next decade of my life.
Waveforms
The problem with reproducing nature by means of microphones and a tape recorder is, simply put, a combination of inaccuracies. Recording technique is at best an art and hardly perfection. Since most recordings are of the human voice and musical instruments, people are willing to "fill in the spaces" with their imagination, even though the reproduction hardly ever matches the sound of real-life occurrences.
Since distortion is inherent in each element in the chain of acoustic reproduction, the net product is a result of compromises that may be pleasing but are seldom accurate. The first problem is that the sounds of nature are not merely a specific set of frequencies, such as the human voice or a particular musical instrument, but are often a form of "noise" that contains hundreds if not thousands of specific frequencies, all of which contribute to the makeup of the particular sound. The second problem is that of reference: hardly anyone has heard Mick Jagger without a microphone, but almost everyone has heard a rainstorm or the sibilant expiration of an ocean wave and knows what the real sound should be.
Into this maelstrom of inaccuracy I plunged with my trusty Uher portable stereo reel-to-reel tape recorder and a tangle of microphones and cables. But the inaccuracies of my highly regarded professional equipment continued to prevail, and nearly a year later I had produced a hundred stereo recordings not one of which actually sounded, to my mind's ear, like the ocean I wanted to hear.
Synthesized Surf
I had a number of friends who were actively involved in the arcane rites of computerdom, and one of them took particular delight in hearing of my analog recording travails. "Digitalize, my friend!" he was wont to say. "Bring your ocean to me, and I'll save you grief."
Well, since I had a shelf full of bad tapes, who was I to argue with destiny? This fellow had worked in a famous East Coast lab for several years, developing a proprietary speech synthesis program for the phone company. His program broke down spoken words into microsecond increments, analyzed each segment digitally and created new words by applying a mathematical formula based on various parameters of human speech. Quite frankly, I doubted if anything would come of our experiment. But I love computer rooms, and his was near state-of-the-art (at least by late sixties standards): an IBM 360 with all the bells and whistles.
We started with a loop from my Brighton Beach tape, which was only a few minutes long. This proved ideal for the computer, since the program we were using limited broadband audio input to approximately two minutes. The first few tries yielded little more than noise from the monitor speakers as we adjusted such technical niceties as I/O parameters, dynamic range and a random number generator to interface with selected waveforms.
With the hands of the wall clock well past midnight, it seemed apparent that nothing would come of this quixotic venture. Then suddenly we both grew still and listened attentively to the output of the monitor speakers. Rolling out through the grille cloth was a beautiful, tranquil ocean sound I had never heard before. The splice on the loop we were using for input could not be detected, as an electronic random noise generator reprogrammed the waveform parameters with each cycle and created subtle new waves that never repeated. By adjusting bandwidth constraints, we got the sound to grow more and more realistic until what we heard was a serenely majestic ocean sound complete with bubbling surf and a faintly perceived, eerily synthesized foghorn.
By this time it was near dawn, so we decided to pack it in for the day. The next night I brought a variable speed instrumentation-type recorder, and we improvised a delay to simulate variable stereo separation by subtly shifting from left to right on a continous basis to eliminate "windowing" of the channels. After a few experiments we found that, by adjusting input and output tape machines to a very slow speed, our tape was able to track the signal almost exactly. At this speed, recording a tape for half-hour playback would take eight hours of computer time, so we set everything up, punched the record button and spent the rest of the night drinking coffee and munching greasy hamburgers at an all-night diner.
By dawn, when we signed back in, the tape had run out. We rewound and sat back. Not only was the recording superior to anything I'd ever done "in situ," but it had such an astonishing bandwidth and dynamic range that it could be played at different speeds and still sound exactly like a "perfect" ocean in completely convincing stereo. We had, it seemed, created the first digitally produced broadband recording.
Roll-Out
I took copies of the tape, which I had jokingly titled The Psychologically Ultimate Seashore, to a psychology professor out on Long Island. I wanted him to try them out on his graduate students as a potential aid to concentration and relaxation. A week later I got a call from him. He had done double-blind testing during sleep research and found that the subjects had had quite vivid dreams after listening to the tapes. Many of the subjects had reported that they felt unusually refreshed upon awakening. In addition, experiments utilizing difficult reading matter had shown that comprehension and reading speed in some instances had doubled when the ocean was played in the background. More importantly, the participants had actually been upset when the tape ended, and a few had even pleaded for copies for their personal use.
Several months later Environments Disc One had been hastily packaged, complete with comments on the back of the jacket detailing the computer synthesis, the glowing reports of the test subjects and instructions on how to play the record. A test market at the Harvard Coop had been arranged, and the ocean outsold the Beatles, especially at exam time. Articles appeared in Newsweek and the New York Times, comparing the record's effects to those of soma, Aldous Huxley's imaginary drug described in Brave New World.
My obsession had become reality and a new form of recorded sound had been born via computer. I sold the rights to Environments Disc One to Atlantic Records and retired for a while. Mother nature had been digitalized.
Return to Table of Contents | Previous Article | Next Article
