Keunwoo Choi

I also hadn't had any experience until a couple of months ago, and actually it was really good experience PR-ing some stuffs. There are unfamiliar aspects that you should know/follow...

Is A-weighting good enough for you? Otherwise you can take a look at https://github.com/keunwoochoi/perceptual_weighting, which is a perceptual weighting code based on ISO226. @bmcfee I once was thinking about PR...

> Different weighting curves? Yes, curves depends on frequency *and* loudness. It is tricky to match the physical energy and energy computed in digital, but we can assume the full-scale...

Hi, yes for sure, but naturally with CNN sequential keras model. You'd want to study how to use Keras **Functional API** to have multiple branches and concatenate them later. If...

sounds great, thanks!

Hi, first of all, your understanding about the current implementation is correct. But I set it intentionally and named it `magnitude_`, instead of either `amplitude_` or `power_`. So it's not...

Thanks all, and please keep commenting. But let me confess. I started to use Pytorch as a main DL library for my work and there are less time I can...

@Yannik1337 Sure. TF doesn't tell me where is the error exactly, especially when it comes to a real experiment that involves tfrecords, tf.data.Data, customized metrics and loss functions, data preprocessing,...

Hi, first of all, I don’t know. I’ll guess a bit. A direct/automatic application of full integer can be dangerous since the dynamic rance of melspectrogram magnitude (before decibel scaling)...

Hi, could you double check the installation? And where is the python executed?