The speech that's generated still sounds relatively artificial. This is in comparison with Adobe's tech where it samples a real human voice and allows speech t

The speech that's generated still sounds relatively artificial. This is in comparison with Adobe's tech where it samples a real human voice and allows speech to be inserted (http://arstechnica.com/information-technology/2016/11/adobe-voco-photoshop-for-audio-speech-editing/ http://arstechnica.com/information-technology/2016/11/adobe-...).