Prix

The Prix Ars Electronica Archive is a collection enabling search and viewing of all the submissions since 1987. The award-winning projects are documented with catalogue texts and audio-visual media. All other submissions can be searched by title/artist and displayed with year, category in list form. Please cite the credits (artwork name, artist and photographer) and only use the materials if your article is related to Ars Electronica. 

Digital Musics & Sound Art Anerkennung - Honorary Mention 2021

Recurrent Morphing Radio

Interspecifics
Over the last few years, access to music via streaming platforms has become not only virtually endless but also highly personalized. Behind every sound discovery engine is a sophisticated commercially-oriented formula carefully designed for inducing a long-lasting manufactured demand. Algorithms that seem to know you well enough to decide on the perfect suggestion for what you should listen to next. Recurrent Morphing Radio (RMR) is a generative neural audio production system fueled by the most consumed genres on digital music platforms. The system extracts and restructures the main audio features found in these styles and generates new sounds that are later continuously broadcasted. RMR is an endless music production machine in which over-consumption equals over-training. A living metaphor of a cultural process intervened by a capitalist market with homogenizing effects on the aesthetics of music.  

The result is a piece of generative sound art, produced by a meta-creative agent, in which the materiality of the original sample dissolves to such an extent that turns into noise over time. For the development of RMR, Interspecifics have employed computer models such as  Recurrent Neural Networks (RNN) to build an automated system using platforms such as OpenAI’s Jukebox, SampleRNN, and Wavenet, among others to deconstruct musical structures according to their original characteristics. The most popular sounds serve as a basis for training the system into generating new materials, which are, in turn, subjected to the same treatment repeatedly, hence its name. Apart from continuously producing music, the system generates the titles and also selects the order in which every song and list is played. It does it by using Markov Chains, a type of discrete stochastic process in which the probability of an event occurring depends only on the immediately preceding event. All the tracks produced by the system are available for download from the project database.

Links: https://int-lab.cc/rmr/, https://int-lab.cc/rmr/database/
Project development by Interspecifics: Leslie García, Paloma López, Emmanuel Anguiano, Felipe Rebolledo 
Collaborators: Diego Montensinos, José Antonio Salinas Hernández 
This piece was commissioned by the Haus der Kulturen der Welt for the Festival The Disappearance of Music, curated by Zuri Maria Daiß and Detlef Diederichsen.
Interspecifics (INT) is an independent artistic research bureau founded in Mexico City in 2013. We have focused our research on the use of sound and A.I., to explore patterns emerging from biosignals and the morphology of different living organisms as a potential form of non-human communication. With this aim, we have developed a collection of experimental research and education tools we call Ontological Machines. Our current lines of research are shifting towards exploring the hard problem of consciousness and the close relationship between mind and matter, where magic appears to be fundamental. Sound remains our interface to the universe.