PyMasVis is a reimplementation of MasVis in Python.
Analyze a file by running python src/analyze.py filename [filename ...]. The result will be output to filename-pymasvis.png. Spotify analysis is done by running python src/analyze.py -u username -p password spotify_uri. Result when using Spotify is output to the current working directory.
There is also a web app which enables you to use a webbrowser as a UI. Start it with python src/webapp.py and point your browser at http://localhost:5000 it is possible to change the listening address and port with --host and --port.
The stand alone analyser requires the following components
- FFmpeg
- NumPy
- SciPy
- Matplotlib
- Pillow
The web app requires
- flask
- flask-socketio
- eventlet
- pyzmq
Systems with a package manager might have the required packages available through it. If that is not possible then installation using for example pip might be possible.
For standard usage:
pip install numpy
pip install scipy
pip install matplotlib
pip install pillow
To analyze tracks from Spotify:
pip install pyspotify
PyMasVis suppports all files that FFmpeg supports as PyMasVis uses FFmpeg to convert the file to raw PCM data before analysis.
Histogram is calculated using a maximum of 2^18 bins regardless of real bit depth. The "bits" result is scaled to the presumed real bit depth.
Scipy requires ATLAS/BLAS/LAPACK. Fortran compiler (gfortran).
PySpotify might require a separate installation of libspotify. Libspotify can be downloaded from https://developer.spotify.com/technologies/libspotify/
On Linux the contents of lib/ goes in /usr/local/lib and include/ in /usr/local/include
On OS X libspotify.framework/Versions/XX.XX.XX/Headers/api.h goes into /usr/include/libspotify and libspotify.framework/libspotify is copied to /usr/lib/libspotify.so