The role of an interactive music system (IMS) is to accompany musicians during live performances, like a real musician. It reacts in realtime to audio signals from musicians, according to a timed specification called mixed score, written in a domain specific language. Such goals imply strong requirements of temporal reliability and robustness to unforeseen errors in input, yet not so much studied in the computer music community.We present the application of model-based testing techniques and tools to a state-of-the-art IMS, including the following tasks: generation of relevant input data for testing (including timing values) following coverage criteria, computation of the corresponding expected output, according to the semantics of a given mixed score, black-box execution of the test data and verdict. Our method is based on formal models compiled directly from mixed scores, and passed, after conversion to timed automata, to the model-checker Uppaal. This fully automatic approach has been applied to real mixed scores used in concerts and the results obtained have permitted to identify bugs in the target IMS.

We present a visual notation framework for real-time, score-based computer music where human musicians play together with electronic processes, mediated by the Antescofo reactive software. This framework approaches the composition and performance of mixed music by displaying several perspectives on the score's contents. Our particular focus is on dynamic computer actions, whose parameters are calculated at run-time. For their visualization, we introduce four models: an extended action view, a staff-based simulation trace, a tree-based hierarchical display of the score code, and an out-of-time inspector panel. Each model is illustrated in code samples and case studies from actual scores. We argue the benefits of a multifaceted visual language for mixed music, and for the relevance of our proposed models towards reaching this goal.

The role of an Interactive Music System (IMS) is to accompany musicians during live performances, acting like a real musician. It must react in realtime to audio signals from musicians, according to a timed high-level requirement called mixed score, written in a domain specific language. Such goals imply strong requirements of temporal reliability and robustness to unforeseen errors in input, yet not much addressed by the computer music community.We present the application of Model-Based Testing techniques and tools to a state-of-the-art IMS, including in particular: offline and on-thefly approaches for the generation of relevant input data for testing (including timing values), with coverage criteria, the computation of the corresponding expected output, according to the semantics of a given mixed score, the black-box execution of the test data on the System Under Test and the production of a verdict. Our method is based on formal models in a dedicated intermediate representation, compiled directly from mixed scores (high-level requirements), and either passed, to the model-checker Uppaal (after conversion to Timed Automata) in the offline approach, or executed by a virtual machine in the online approach. Our fully automatic framework has been applied to real mixed scores used in concerts and the results obtained have permitted to identify bugs in the target IMS.