Search the Community
Showing results for tags 'database'.
I'm ok (not great) with music and algorithms, but I know *nothing* about Lisp, so please excuse me if I'm completely off base... I'd like to set up a system where I can tag parts of midi files and then use random weights to create new pieces that are similar to the source material. For example, I'd like to create a new melody from the openings of three different pieces where each of the source motifs are tagged with something like "dark" "rhythmic" "complex" In my day job the way I'd do this is to have all the motifs, melodies, basslines, chord progressions, descriptive tags, etc. stored in a database, and I'd run a query: "SELECT * from motifs WHERE tag LIKE 'dark' LIMIT 3. And then I'd basically use random numbers to select which of the resultant notes would play and when. I'd then rate the result, and if it's good enough INSERT it back in to the database to provide a new generation of material So my question is, would I need a database in OpusModus, or is there a "Lispy-way" to do this kind of genetic manipulation? If I need a database, any recommendations? Thanks for your patience!! Lance