Keunwoo Choi

Hi, yes I removed the layer intentionally. It was a part of design choices. You can consider similar stuff like instancenorm (https://www.tensorflow.org/addons/api_docs/python/tfa/layers/InstanceNormalization).

Hi, sorry for a late reply. In a preference of using `tf.signal.stft`, I removed it in the newer versions of `Kapre`. To be honest, I miss the feature ever since...

Hi, Sorry for the confusion. It was removed in `0.3.0`. (https://github.com/keunwoochoi/kapre/pull/81/files?authenticity_token=vUHWbAmgbdztV8SeUSdRMmTYVooJVXPA5Rnm8DEew7FQW2yb0Gf0c8hxScmybKC7n0RMEOWsTHmP%2Fn706W1fRw%3D%3D&file-filters%5B%5D=.ini&file-filters%5B%5D=.ipynb&file-filters%5B%5D=.md&file-filters%5B%5D=.py&file-filters%5B%5D=.sh&file-filters%5B%5D=.npy#diff-0d0a9affae3a2c387b9e3a7806e3abe798d59dce44dee23670499c3dd9690d72)

- [x] mfcc? - [ ] time-stretch? - [ ] frequency shift?

- [x] fix test for phase

- [x] energy - [ ] loudness - [ ] HPSS

- [x] tf.signal.frame - [x] mu law - [ ] pcen - [ ] freq and time mask (spec aug) - [x] augmentation - swap channels - [ ] augmentation...

- [x] frequency-aware conv layer? (@kkoutini any chance you'd be interested in making a PR?)

- [ ] `InstaneousFrequency(pad_end=True)` - [ ] pre-emphasis, de-emphasis - [ ] LPC - [ ] ERB - [ ] LFCC

- [x] change numpy requirements to 1.18.5 to be compatible with tensorflow. - [x] enable model name for composed layers