Rating is available when the video has been rented.
This feature is not available right now. Please try again later.
Published on Sep 9, 2016
This is the result of running DeepMind's synthesized vocal utterances through YouTube's automatic caption generator.
[[ make sure you enable closed captioning by clicking on the "CC" button in the bottom right of the control panel ]]
Google's DeepMind first described WaveNets, deep neural networks capable of generating synthetic speech and other kinds of audio, in a blog post on Sept 8, 2016: https://deepmind.com/blog/wavenet-gen...
WaveNets can synthesize speech conditioned on a specific text sequence, or unconditioned, as in these six samples from the original post, where it just sounds like pseudo-English babble. What are they actually saying? Let's run them through YouTube's automatic captioning system to find out.