[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: IDyOM v1.1



Hi Marcus 

This is amazing! It takes me back to some tangential work done by Max Mathews with the radio baton, that later went on to the Marimba Lumina (Infact had the pleasure of seeing Lucas Ligeti play last year in India)

Which DSP stack did you use to start off with? And the ML Library?
Would be nice if you could upload the source code, or even just the modified GNU Libraries!

Looking forwards
Regards
Sharath

On Mon, Oct 6, 2014 at 4:19 AM, Marcus Pearce <marcus.pearce@xxxxxxxxxx> wrote:
Dear All (with apologies for cross-posting),

I'm pleased to announce v1.1 of IDyOM (Information Dynamics of Music) - a framework for constructing multiple-viewpoint, variable-order Markov models for predictive statistical modelling of musical structure. The system generates a conditional probability distribution representing the estimated likelihood of each note in a melody, given the preceding musical context; it computes Shannon entropy as a measure of uncertainty about the next note and information content as a measure of the unexpectedness of the note that actually follows.

Updates in this release include greater control of caching behaviour, easier installation, improved internal representations of music objects and fixes to Lilypond export.

The software and documentation are available at:

https://code.soundsoftware.ac.uk/projects/idyom-project
Marcus

--
Lecturer in Sound and Music Processing
Queen Mary, University of London
Mile End Road, London E1 4NS, UK
Tel: +44 (0)20 7882 6207
Web: http://webprojects.eecs.qmul.ac.uk/marcusp
Lab: http://music-cognition.eecs.qmul.ac.uk/