User Tools

Site Tools


sarroff

deepAutoController

Authors Andy Sarroff, Sean Manton, and Phil Hermans
Affiliation Dartmouth College, Boston University, and Dartmouth College
Code here at github!!

When we build deep graphical models, it is difficult to gain intuition about the contributions of individual neural units or groups of neural units. This project aims to achieve two goals:

  • Unordered List Item Allow a user to gain a better understanding of the code layer of a deep autoencoder
  • Create new sounds by performing sample-based synthesis using a deep autoencoder.

The code in this repository will allow one to

  • Collect a corpus of Constant Q Fourier Transform features
  • Train a deep autoencoder having an arbitrary number of hidden layers and units per layer
  • Perform music by modifying the optimized model interactively with a midi controller

Developed by Andy Sarroff (http://www.cs.dartmouth.edu/~sarroff) with help from Sean Manton and Phil Hermans.

Full source available at github: https://github.com/woodshop/deepAutoController

Project pitch slides

sarroff.txt ยท Last modified: 2013/06/30 17:19 by sarroff