WaveNet: A Generative Model for Raw Audio
WaveNets are able to generate speech which mimics any human voice and which sounds more natural than the best existing Text-to-Speech systems, reducing the gap with human performance by over 50%.
Source: deepmind.com