search

chrisdonahue / wavegan

WaveGAN: Learn to synthesize raw audio with generative adversarial networks
MIT License
1.32k stars 283 forks source link

Add Generate Script #29

Open robdmac opened 5 years ago

robdmac commented 5 years ago

This wraps the audio generation into a script and generates an audio file reminiscent of (and to make comparisons with) the tensorflow-wavenet repo: https://github.com/ibab/tensorflow-wavenet/blob/master/generate.py

chrisdonahue commented 5 years ago

Thanks for the PR and sorry it has taken me a while to get back to you. I like the idea of having this generate script, but it would make sense to also have one for SpecGAN. Can you possibly rename this to wavegan_generate.py and add a corresponding one for specgan? If that is too much effort, I can leave this PR open and get around to writing the code.

Also would be nice to add an argument specifying the number of samples to generate.

itsKaspar commented 5 years ago

ej, i'm having issues with the generating script i've installed wavenet through pip and pip3 but when I run the script through P3 I get

ImportError: cannot import name 'WaveNetModel'

so I suppose the code isn't for P3 and when I run it through P2

ImportError: cannot import name ABC

and I can't seem to find a way to install abc which package is this refering to ?