Let's talk about Google DeepMind's Wavenet! This piece of work is about generating audio waveforms for Text To Speech and more. Text To Speech basically means that we have a voice reading whatever we have written down. The difference in this work, is, however that it can synthesize these samples in someone's voice provided that we have training samples of this person speaking.
__________________________
The paper "WaveNet: A Generative Model for Raw Audio" is available here:
https
آی-ویدئو