Skip to content

cjam/depth-sounder

Repository files navigation

Build Status

depth-sounder

Neural-Network-Based Audio Synthesis

Getting started

My Setup: - OSX
- Python 2.7.10 installed - Pycharm as IDE (got an education license, but I don't think its very expensive)

Prerequisites:

sudo pip install virtualenv

For sake of simplicity I'm going to commit the Virtual Python Environment without its dependencies installed. Hopefully this will help speed up co-development as all you will have to do to get up to speed is:

cd project_dir
sh setup.sh

the setup.sh file has all of the weird steps included for setting up your python environment

After running setup.sh you should now have an environment that can be used for the project.

The web.py is a simple web application built on Flask + Flask-socketio which will facilitate real time configuration of the underlying neural net processing.

Web Clients

I've typescript-ified our web clients and placed the code in /web/clients/

There are a few pre-reqs to getting this to work:

  • Install npm
  • npm install -g tsd

In order for typescript to properly compile these, you must run the following command in this directory:

tsd install

I'll get around to adding it to the setup scripts.

Scripts

setup.sh

Sets up your environment initially to be able to run the python scripts

install.sh

Installs a package into the environment and exports the list of dependencies to requirements.txt in the root directory

uninstall.sh

Uninstalls a package from the environment and exports the list of dependencies to requirements.txt in the root directory

bundle.sh

Exports the list of dependencies (excluding pyAudio, see Appendix T)

Appendix T*(roubleshooting)*

I Can't use matplot lib!

if your on OSX you'll get an error talking about python being installed as a framework. To get around this, run your script using sh frameworkpython.sh myscript.py

This will run the python shipped with OSX but set the home directory to that of our virtual environment.

PortAudio on install

There was an issue with installing PyAudio (missing PortAudio.h) since we're using our own virtual env:

Here's how it was solved:

brew install portaudio
pip install --global-option='build_ext' --global-option='-I/usr/local/include' --global-option='-L/usr/local/lib' pyaudio

The rationale behind this is that we're using our own virtualenv for our project but portaudio gets installed into the /usr/local/lib so we need to link to the machine folders when building it on install

About

Neural-Network-Based Audio Synthesis

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published