VRA_peek is an audio-visual performance, in which sound is created exclusiveley out of images. The performer uses code to produce images, which are converted in realtime into a stream of audio-data, creating a totally in-sync audio-visual experience. A squared screen projection, shows the current visual information, being also the source of the hearable audio.

The generated image is 1024x1024 pixels. One horizontal row of pixels, which is highlighted green, is read out with 32hz from left to right. This horizontal reading speed is dictated by a sawtooth oscillator and can be alterded. For this project however, the oscillator operates at a fixed rate giving the base frequency of 32 hertz. This horizontal row of pixels can be offset along the vertical axis by hand or automation, choosing which part of the image is read out.
In this way, the pixel data is directly transformed into an audio-waveform, which means one pixel represents one sample of audio.

The images contain mostly simple elements, like stripes, gradients or circles, which then can be distorted by, or mixed with noise. This way the character of the performace reaches from long-drawn, slow moving sinewave harmonics, over sharp concrete squarewaves, to complex evolving rhythmical structures, as a result of hectic stroboscopic shapeshifting.
As a consequence of the audio being sourced out of the image data, every visible change is instantly perceptible at the audio layer.

First demonstration (work in progress/early stage of the project/poor quality):
https://www.youtube.com/watch?v=X1Z3tSUNFMY

High quality recording will follow.