Chris Hold

Hi! Thanks for your comment! This package is mainly focused on research / development purposes, and you are right I kind of assume some expertise in audio signal processing. What...

> I would like to set up a scene with the camera (I saw that in one of your exemples I think), then 2 sound sources (more sources actually but...

I could also make it fail locally, however, only a single time so far. Seems to be a lot less frequent when executed on a local machine. Yes, it seems...

Another traceback: ``` Test on ubuntu-latest with Python 3.7 =================================== FAILURES =================================== _____________________________ test_render_bsdm[2] ______________________________ test_jobs = 2 @pytest.mark.parametrize('test_jobs', JOB_COUNTS) def test_render_bsdm(test_jobs): sdm_p, sdm_phi, sdm_theta = [*np.random.randn(3, 1000)] hrirs =...

Hm, I am using `multiprocessing.Array`, which already includes a lock... https://docs.python.org/3/library/multiprocessing.html#multiprocessing.Array I am quite surprised why this particular function seems to fail while others don't.

Seems it was indeed the `+=` inplace operator. I'm also casting back to a numpy array, so the operator seems to be not thread safe there. cf77fa8f9065e7ecca2f8b6667baca8fdd42392b and f5e5654b7c463710f09c92ca4c07a272c32c0a8a target...

Re-opening, because still not resolved. The last times failing seem to happen always on Windows. Traceback: ``` Cross-Platform Test / Test on windows-latest with Python 3.6 test_jobs = 2 54...

Hi! What function is this about? In general, the sources are the emitters, and the listeners are the receivers.

Hi Benjamin, Thanks for your contribution! In fact, you discovered a bug. The in-phase weighting should be equivalent with a higher-order cardioid weighting, defined in https://github.com/chris-hld/spaudiopy/blob/e463859c24072a749e01d952a60b1b11196bcb79/spaudiopy/sph.py#L828 However, it seems there...

I would propose to merge your PR, and to base `cardioid_modal_weights` on the in-phase weighting coefficients, normalised for unit amplitude in steering direction (the 4pi factor). Is that ok with...