MusicVisualizer is a project written in Python that visualizes mp3s and outputs mp4 videos. This is currently in alpha development but is showing promise, executing it's main goal of visualizing music.
Currently our install script ain't perfect. First, make sure you have pip3
installed with
sudo apt-get install python3-pip # Get pip3
sudo pip3 install --upgrade pip # Get the latest version of pip
sudo apt-get install ffmpeg # Does stuff with mpegs
sudo pip3 install numpy # Numpy has to be installed seperately
sudo setup.py install # Run the install script
You should be good (hopefully). Additionally there is a setup script setup.sh
that does a lot of the legwork for you (in fact, a simple call to setup.sh
should be everything you need on an Ubuntu installation). However, if for
whatever reason this doesn't work (and you don't feel like fiddling) you can
run this on a
sanitized install of Lubuntu
through Virtual Box. The install has credentials are user/password and
in-depth notes on installing and running are included on launch. Also worth
mentioning is that the virtual box has a disk size of 16GB and currently we are
writing temp files to /tmp/ and not releasing them; on a small hard drive like
this we can easily run out of room. This is a bug and an issue has been
registered but as of now it is not fixed.
There are two ways to run MV; the first is through a command line interface, and this gives you the maximal flexibility. The second is through the GUI and this gives restricted access.
Assuming that we installed everything correctly, we can run visualize input.mp3 and after some crunching an output.mp4 will appear in the directory
that you called visualize from. Additonally you can pass an output argument
-o path/to/destination/file.mp4 to specify where you would like the output to
end up. For more information run visualize --help.
Run src/mvp and you should be good!
We are using the Python module unittest for all of our testing needs. From the
src directory issue a
python3 -m unittest discover # Discover all tests in the project
and this will automatically perform all tests. The testing infrastructure is still scant but is growing by the hour!