Skip to content

Music-Machine-Learning/IMprovEEsation

Repository files navigation

#IMprovEEsation

Intelligent Musical Evolutionary Entertainment

##Live Demo

IMprovEEsation patterns

The real musicians play their instrument upon their musical taste, their known musical expierence and the real time feeling given by the other musicians in a live session. In a similar way, we tried to represent these behaviours basing our software in a "patterns" perfomance mechanism. A "pattern" is defined as a collection of measures. A measure, in musical notation, is a segment of time defined by a given number of beats, each of which are assigned a particular note value. By analogy, we used a similar model to define a measure, but unstead of assign a single note per each beat, we assigned a set of probabilty values per each note for a single beat. This gives us the opportunity to model the wish of a certain musician to improvise over some pre-defined patterns, which ones may depict a musician past knowledge. Moreover, to model the taste of a musical sequence, we introduced a "tags" system through which every measure collects a set of words that defines its dynamic. In this way a musician may choose different patterns according to a pre-choosed genre or according to the playing dynamics of the other musicians. To understand more on how the patterns are defined and used, check the related file inside the "Documentation" folder.

Setup

TODO

About

Intelligent Musical Evolutionary Entertainment

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 4

  •  
  •  
  •  
  •