NOISA: Network of Intelligent Sonic Agents

Research output: Artistic and non-textual formExhibitionSolo art productionpeer-review


NOISA, the Network of Intelligent Sonic Agents, is an interactive music system that monitors the performer’s actions and provides autonomous and non-intrusive counteractions. In this co-creative music performance, intelligent sonic agents are designed to support the music performance by providing responses that encourage and maintain the communication and motivation of the performer with the NOISA system. The sonic agent is an entity in each NOISA instrument that learns relationships between sonic agents, performer and musicscape. Each sonic agent has a machine-learning model for predicting and adapting its behavior during the performance. Regression algorithms, which use a tree model to derive the predicted parameter value from decisions and branches, lead to predictable and representative behavior based on the learning data. The NOISA project elevates people’s creative thinking, engages them in investigation, and contributes in positive ways to the co-created arts with machine learning and AI.
Original languageEnglish
Place of PublicationLinz
Publication statusPublished - 6 Sep 2019
MoE publication typeF1 Published independent work of art
EventArs Electronica Festival - PostCity, Linz, Austria
Duration: 5 Sep 20199 Sep 2019


  • artificial intelligence
  • Music technology
  • digital musical interactions

Field of art

  • Contemporary art

Fingerprint Dive into the research topics of 'NOISA: Network of Intelligent Sonic Agents'. Together they form a unique fingerprint.

Cite this