Listening to the radio the day before yesterday, I caught the tail end of a report about a new music machine called The Humanizer. Let me give you a little bit of background.
When a piece of music is played it follows notes put down on paper (mostly!) by a composer. The musical notes and instructions are fairly exact, showing what the composer wishes to be played, with timings, pitch, volume and so on.
When an orchestra plays a symphony, it works under the direction of a conductor who has taken the work to be played and interpreted it according to his or her own desires, within the realms of what the composer has laid down on paper. The interpretation is different according to whichever conductor is performing the works and according to the orchestra itself. No two performances are ever the same, no two conductors have the same ideas. There are constantly small ‘human’ differences involved, a slight delay here, a slightly off-tune note there, mistakes, additions and so on. This is what makes listening – and performing – music so wonderful; the human factor. And this is also why there are so many copies of a single symphony – for example – available. Each one, based on the same work, is essentially different.
Scientists – or computer programmers for all I know – came up with a machine which can produce a composer’s work exactly according to the timings and instructions that he or she laid down upon publication. Each time this work is played it is exactly the same, no differences from one performance to the next with no individuality whatsoever. Theoretically, only one copy of a work is now needed, and everyone can enjoy the same piece of music exactly the way it was written.
However, the same people came to see that the individuality of a piece of music lies in these small differences, be they but a microsecond difference from one another. So someone came up with a standard machine which built in these small differences of timing, volume and so on, all to a set formula.
This, in turn, meant that a piece of music could be played in several different ways, according to the differences programmed into the software, but the differences would always be the same.
Not to be outdone, the Max Planck Institute has now come up with an even better model, whereby the differences are random, not set and, therefore, more human. A machine can now produce a piece of music in many different varieties according to the listeners’ desires.
In effect, they have worked through a series of programs to turn a perfect piece of music, as the composer intended, into a machine-produced piece of music imitating a human-produced piece of music.
Wouldn’t it have been just as easy to let humans carry on playing in the first place?
Love & Kisses, Viki.