Subject: Re: Modeling musical beginning From: Pablo Hernan Rodriguez Zivic <elsonidoq@xxxxxxxx> Date: Wed, 23 Sep 2009 10:00:31 -0300 List-Archive:<http://lists.mcgill.ca/scripts/wa.exe?LIST=AUDITORY>Hey, Thank you! I'll try to get that book. Scott Spiegelberg wrote: > Hi Pablo, > > Look at David Cope's work with EMI (Electronic Musical Intelligence). > He would create databases of scores in a particular genre, like Chopin > Mazurkas, that EMI would use to create a new mazurka. > With this case, and any case of composition, the starting pitch will > depend a lot on the style of the composition. If it is a tonal piece > in the style of the Classical period, it is likely to start on the > tonic, mediant, or dominant. Romantic pieces often start on the > dominant, leading tone, or supertonic as much as the tonic triad > notes. And atonal pieces can start on any pitch. > > Scott > > On Sep 22, 2009, at 10:13 AM, Pablo Hernan Rodriguez Zivic wrote: > >> Hello everyone! >> >> I don't know if you remember me, I'm the crazy one trying to make the >> computer compose melodies =D >> >> I'm going very fast and it seems that I'll finish my grad thesis >> soon... however I'm kind of stuck with one detail: >> The model that I'm using for the melodic contour, needs another >> model for the musical beginning because it's a conditional model >> based on the lasts notes that have been played. >> >> I've heard of a study that analyzes a lot a scores in order to find >> some pattern in how music begins. >> >> Does anyone know about that? >> >> Thanks a lot!! >> >> Pablo >