raymondxyy / pyaudlib

A speech signal processing library in Python with emphasis on deep learning.
MIT License
31 stars 6 forks source link

SSF #30

Closed patconrey closed 4 years ago

patconrey commented 4 years ago

High Level

This PR finishes the existing implementation of SSF.

Details

Previously, we were just returning the ratio of processed power to input power. Now, the algorithm will also perform gammatone weighting. This PR also updates the functionality of the SSF enhancer.

Lingering Questions

NB

This is based off the old ssf branch. A lot has changed in the repo since then. Some of the functionality here may already be abstracted into other functions. Please let me know! Along those lines, we may not want to merge in some of the other changes on this branch that aren't directly related to the SSF algorithm. Please look over those changes too, as I'm not sure what they were intended to do.

andrewwuan commented 4 years ago

For PR template, I suggest https://github.com/stevemao/github-issue-templates/blob/master/questions-answers/PULL_REQUEST_TEMPLATE.md.

If this looks ok, I can add it as our default PR template.

raymondxyy commented 4 years ago

The PR template looks good, @andrewwuan go ahead adding it.

codecov[bot] commented 4 years ago

Codecov Report

Merging #30 into master will increase coverage by 1.08%. The diff coverage is n/a.

Impacted file tree graph

@@            Coverage Diff             @@
##           master      #30      +/-   ##
==========================================
+ Coverage   39.18%   40.26%   +1.08%     
==========================================
  Files          40       40              
  Lines        3529     3561      +32     
==========================================
+ Hits         1383     1434      +51     
+ Misses       2146     2127      -19     
Impacted Files Coverage Δ
audlib/sig/fbanks.py 68.68% <0.00%> (+1.64%) :arrow_up:
audlib/enhance.py 15.26% <0.00%> (+15.26%) :arrow_up:
audlib/noise.py 12.50% <0.00%> (+12.50%) :arrow_up:

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update 2516080...8c892be. Read the comment docs.