Noise Curve is a sound (and visual) installation that searches for hidden information in almost silent sound samples. The audio samples come from field recording in uninhabited places.
Using a custom software, the sound samples are analyzed with a Fast Fourier Transform (FFT) algorithm, that produces the frequency domain of the signal. These frequencies are used to plot a curve on the screen that leaves an indelible mark over time. This curve is used to represent the sound in a single two-dimensional image.
Noise curves are a common way to measure and specify background noise in unoccupied buildings and spaces. Their purpose is to produce a single value representation of a complete sound spectrum. From Wikipedia
Materials: Mini projector LCD, Rasperry Pi, custom software, headphones, sample sounds, Kodak Carousel S-AV 2010.