Brian McFee
Brian McFee
**Is your feature request related to a problem? Please describe.** We currently have two interfaces to ingesting audio: `load` and `stream`. Aside from the obvious functional differences between the two,...
Note: this should be among the last issues implemented in 0.10 because of how our doc build CI works. When we release 0.10, we should drop the historical build of...
Opening a conversation here to plan out how we want to handle deprecating and removing audioread as a dependency. ## Background When librosa development began, audioread was the best option...
**Is your feature request related to a problem? Please describe.** I often find myself needing to look up frequency ranges (`fmin, fmax` in Hz) for various sources, e.g., human voice...
The visualization of IIR filters that we use here is cluttered and difficult to parse.  I think we could do much better with something more like a...
Our doc site has a [reimplementation](https://librosa.org/doc/latest/auto_examples/plot_vocal_separation.html#sphx-glr-auto-examples-plot-vocal-separation-py) of the repet-sim method for foreground/background separation, which gives a nice demo of a few features: - audio source separation and resynthesis - non-local...
**Is your feature request related to a problem? Please describe.** At present, the recurrence_matrix function provides a rudimentary bandwidth estimator, which uses the median k'th nearest neighbor distance. A user...
The documentation site is generally pretty navigable, but I think the core submodule (sub-heading "Core IO and DSP") is both too large and at this point, inaccurately named. At present,...
I don't have any clear ideas for this, but I suspect that our main website, front page of the documentation, and project readme could be a bit more inviting (and/or...
**Is your feature request related to a problem? Please describe.** #1018 implements variable-Q transforms. This issue exists to factor out the related methods for inverse-VQT and VQT phase recovery. These...