Hartmann
Neuron Mind in the machine |
||||||||||
The Neuron aims to bring a "human" touch to synthesized sound creation. The sounds are produced by means of a neural network engine, which assigns individual parameters to any given sound. At first sight, Neuron looks like any other keyboard synthesizer, except for the peculiar controls: there's no traditional knobs, but rotary wheels, a whole bunch of displays, and joystick controllers in a translucent orange colour instead. Under the hood, Neuron resembles a contemporary personal computer: it features a powerful processing unit, a hard disk, and memory galore. The instrument obviously needs a lot of calculating power to carry out its task: to analyze and alter source sounds by means of a neural network engine. A virtual neural network is in fact a computer algorithm that mimicks the human brain, and represents a kind of artificial intelligence. This might sound futuristic, the theory behind the neural networks however dates back to the experiments of American scientists Warren McCulloch and Walter Pitts in 1943. The primary task of a neural network is to emulate human pattern recognition skills and to distribute knowledge throughout the network. This way, the neural network "learns" to adapt to various external situations. Normally, this technology would be used to simulate and analyze market conditions or discover clusters in databases. German synthesizer-designer Axel Hartmann used the technology to build what he calls "a breakthrough" in the history of the synthesizer. Great words indeed, but Hartmann obviously means it. He is almost a star in German synthesizer-design: he has designed the interface for Waldorf's famous "Wave" model and Alesis' "Andromeda" - both renowned for their great sound and peculiar looks. For the Neuron, Hartmann has extensively collaborated with Prosoniq software, a specialist in virtual effects. The key technology has been developed in more than ten years of extensive research, and has been tagged "adaptive sound analysis". |
Close-up of Neuron's interface
|
|||||||||
|
||||||||||