http://repmus.ircam.fr/ 2026-04-25T17:34:53+02:00 http://repmus.ircam.fr/ http://repmus.ircam.fr/ttp://repmus.ircam.fr/lib/tpl/repmus/images/favicon.ico text/html 2017-11-08T18:18:03+02:00 http://repmus.ircam.fr/dyci2/demos/cluster_graph?rev=1510161483&do=diff Each agent consists of a knowledge represented by interpolated sub-models trained on a rather large corpus (~50 tunes) and of a intuition represented by a factor oracle constructed on a small corpus (~1 tune). The agents interact by using message passing through a cluster graph with a belief propagation algorithm. This aim to emulate the cognitif process of interplay between musicians, but also how a musician creates their improvisation over several dimensions. text/html 2017-11-08T17:15:28+02:00 http://repmus.ircam.fr/dyci2/demos/corpus_choice?rev=1510157728&do=diff Here, we compare improvisations generated with a Factor Oracle combined with a probabilistic model trained on different corpora. The probabilistic model is an optimised interpolation of smoothed sub-models learnt on a corpus. In the following examples, the probabilistic sub-models used are a bigram on the melody and a model representing the correlations between the melody and the harmony. text/html 2018-03-05T14:21:13+02:00 http://repmus.ircam.fr/dyci2/demos/multi-level?rev=1520256073&do=diff Improvisations on Multi-Level Scenarios We consider as multi-level scenario a description of a temporal structure (e.g. a chords progression) upon several level of hierarchical organisation. In particular, to model a multi-level scenario, we use a phrase structure grammar. text/html 2017-11-08T17:02:36+02:00 http://repmus.ircam.fr/dyci2/demos/probabilistic_fo?rev=1510156956&do=diff Here, we compare improvisations generated with a Factor Oracle alone vs. improvisations generated with a Factor Oracle combined with a probabilistic model. The probabilistic model is an optimised interpolation of smoothed sub-models learnt on a corpus.