Open sewardlee337 opened 6 years ago
Leaving this here for reference purposes: http://python-packaging.readthedocs.io/en/latest/
Per commit dc9f8ee3665b980b001ea546bd39a5b5886aca0c, what I want to be able to do is:
I suck at naming things, so at some point will need help making object and variable names sound better. See sonification-test.ipynb
I'm probably missing something super obvious... is there anything I need to make sure of before running in Jupyter notebook again?
do you have a theme for your object and variable names?
@lj42 Hmmm I don't know off the top of my head and without further investigation. A quick checklist would be...
sound-files
is populated with the .wav
files.pyglet
is installed. (You'd need to run either pip install pyglet
or conda install pyglet
in your terminal.) Not sure if that was super helpful. I'm still trying to figure out how to use pyglet
myself. It just seemed the most promising out of several options...
@lj42 the way I'm thinking of it is...
Can play back the sounds
This is kind of abstract, but is the class of objects an Instrument
? Or Synthesizer
? Or is it a Composition
? Or maybe a Composer
or a Performer
?
I've definitely installed pyglet. Aha, I have not populated the sound-files
folder that'll be it. I'll report back accordingly.
I think the class of objects could make sense instrument
but it's more than that as it's ingesting data too...
Perhaps this is more like a maitre d' or similar as it ingests something (an order) and then outputs something else (dinner) - ah, my vote would be a conductor
what do you think ? The conductor ingests data (the musical score) - maps attributes to members of the orchestra (strings: turn up the volume) and ultimately the instrument or synth plays back the sounds (single musician)
For personal reference:
UPDATE: After a lot of trial-and-error, now figured out a way to play background sound using .play()
method for Conductor
object.
See commit: 1de67bc385614cbb575d91ec36f1b6e73a9bc2fd
MVP Checklist:
import datasong
)
It would be really cool if we can create a Python package like
matplotlib
orseaborn
, but for data sonification rather than visualization. Doesn't have to be super complex at the outset -- just organized and scalable.If we can get this working for 2 or 3 datasets, we can check in with NASA (perhaps via Datanaut Water Cooler Chat) to get general feedback.
To-dos:
Small, disrete tasks may be broken down into smaller issues in Issue Tracker.