zeloe / NoA_I

(Very) Easy ML network

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

NoA_I

Influenced by Arnold Schönbergs dodecaphonism.

Python requirements

  • pip3 install opencv-python
  • pip3 install python-osc
  • pip3 install numpy

Install processing

https://processing.org/

  • install osc5 (external library)

Get MaxMsp

https://cycling74.com/

Download Rave and follow Instructions for training and compiling the external

https://github.com/acids-ircam/RAVE

How to use

  • From Terminal or Visual Studio Code launch NoA_1.py or NoA_1_random.py or both
  • Open the Sequencer_Grid.pde or Sequencer_Grid_random.pde with processing or both
  • Open the MaxMsp patches (use your own RAVE models)

How it works

  • This is an installation, which uses camera input.
  • When it detects a face it starts the actual processing.
  • It saves every 50 x 50 pixel and computes the mean and adds it to the part before.
  • The highest pixel mean will play a sound.
  • The highest pixel mean will be reset to 0.
  • Like this every sound has a higher possibility to get played.

Why...

  • This was part of an university project.

About

(Very) Easy ML network

License:GNU General Public License v3.0


Languages

Language:Max 92.5%Language:Python 5.0%Language:Processing 2.5%