Real world neural networks for interactive music composition in real time

Bickerstaffe, Colin (2017) Real world neural networks for interactive music composition in real time. [Dissertation (University of Nottingham only)]

[img] PDF - Registered users only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (1MB)

Abstract

A new model designed for musicians to use as a tool for practicing and improvisation is developed based on a corpus of Jazz lead sheets. It is lightweight enough to work within the CPU and memory constraints of consumer equipment and operate in real time with input from a standard midi device. The model is also compatible for use on a Raspberry Pi for ultimate portability (at the cost of a significantly longer training time).

Item Type: Dissertation (University of Nottingham only)
Keywords: Artificial Neural Networks, Recurrent Neural Networks, Long Short Term Memory, Musical Instrument Digital Interface.
Depositing User: Gonzalez-Orbegoso, Mrs Carolina
Date Deposited: 09 Jan 2018 09:14
Last Modified: 12 Jan 2018 00:04
URI: http://eprints.nottingham.ac.uk/id/eprint/48526

Actions (Archive Staff Only)

Edit View Edit View