ABOUT SHEEN

ABOUT US

SHEEN AI is an audio-visual AI lab based in Berlin. We are dedicated to exploring and discovering the artistic potential of the latest research in artificial intelligence and beyond, bridging the worlds of science, art and technology. At this intersection, we are creating software and hardware that enables a true connection between audio input and visual output in real-time, enabling a new medium of art. We are a team of AI researchers, creative developers, musicians, visual artists and other visionaries that are interested in enabling the creation of revolutionary, multidimensional experiences.

OUR ACHIEVEMENTS

SO FAR

CAA-Award

Cresc. Biennale 2020: Human Machine

TONALi: I sing the body electric

SHEEN STUDIO is a new media arts studio based in Berlin. We are dedicated to exploring and discovering the artistic potential of the latest research in artificial intelligence and beyond, bridging the worlds of science, art and technology. At this unique intersection, we are creating software and hardware that enables a true connection between audio and visual input in real-time. Our team consists of AI researchers, engineers, creative developers, musicians, visual artists and other visionaries that are interested in creating revolutionary, multidimensional experiences and content.

HOW SHEEN WORKS

Our software combines different neural networks in order to transform sound into visual art. According to the sound (I) the visual AI "paints" visuals onto a dynamic canvas using its imagination (II). The imagination is based on visual concepts that derive from visual reference input, such as images and videos (III), bringing them to life - even live-on-stage.

The Audio AI perceives sound, dissecting it into perceptual auditory components, which are then to be connected to visual features allowing the creation of the desired look and feel of the visual output.

Visual Concepts can be created by providing videos and images as visual reference input (A), or by letting the AI create visual reference input based on text prompts (B). The AI understands and then re-creates entirely new and original output based on the nature of the given visual concept.