sertansenturk / predominantmelodymakam Goto Github PK

even though running the unittests in travis will be tedious because of the essentia, numpy and scipy etc. dependencies, we can still run the nosetest locally

The set should be restricted to certain input type/range for each propery, when applicable

https://guides.github.com/activities/citable-code/

When I run the extractor with a test audio file of 1 seconds, it fails:

results = extractor.run(filename1)
  File "build/bdist.macosx-10.9-x86_64/egg/predominantmelodymakam/predominantmelodymakam.py", line 85, in run
  File "build/bdist.macosx-10.9-x86_64/egg/predominantmelodymakam/predominantmelodymakam.py", line 215, in select_contours
ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all()

I figured out that if the audio is 2-seconds long, it works. So I suspect there is an issue with minimal pitch contour length in melodia.
I would add a more informative error message.