News for hackers. Stuff that matters.
Re:WaveNet: A Generative Model for Raw Audio by dhammack on Thursday September 08, 2016 @05:39PM (#12457578)
I see. I still think it's easier to apply deepmind's feature transform on text rather than to try to invert a neural network. Armed with a network trained on Trump, deepmind's feature transform from text->network inputs, you should be able to make him say whatever you want, right?

Text -> features -> TrumpWaveNet -> Trump saying your text

Re:WaveNet: A Generative Model for Raw Audio by erichocean on Thursday September 08, 2016 @05:42PM (#12457608)
> Armed with a network trained on Trump, deepmind's feature transform from text->network inputs, you should be able to make him say whatever you want, right?

Yes, that should work, and by tweaking the WaveNet input appropriately, you could also get him to say it in a particular way.