This workshop is meant as a hand-on tutorial on using OpenSoundControl (OSC) messaging and signal processing techniques to create interaction between audio and visuals. Our working platforms will be Supercollider and Processing, but the techniques can be easily applied to other programs like PureData, Openframeworks, Cinder, etc.

To extract meaningful information from sound we will work with signals in the Fourier and Time domain. We will look at various descriptors, their theoretical basis and code examples in Supercollider.
In the Fourier domain we can look at information related to the tonal nature of the sound by analyzing magnitudes, centroids and power bands among other descriptors.
The time domain is more appropriate for obtaining rhythmic information, by using onset detectors for percussive attacks, Adaptive Oscillators for beat tracking and also by other statistical methods.

To connect sound to visuals, and viceversa we will demonstrate some examples of how to use OSC message to send information to different programs and different computers either wired or wirelessly.

At the end of the workshop we will see a more complex example of a sound event detector in Supercollider that relays information to a Processing sketch and detonates visuals upon detection.