Google’s DeepMind research lab today published its latest work in the area of speech synthesis, which is better known as text-to-speech (TTS). And the findings are fascinating: A trained version of the new WaveNet artificial neural network produces English and Chinese speech that sounds more natural than Google’s latest implementations of two kinds of TTS systems. […]
DeepMind's WaveNet produces better human-like speech than Google's best systems
Venture Beat - 8 Sep 2016 22:23
Google’s DeepMind research lab today published its latest work in the area of speech synthesis, which is better known as text-to-speech (TTS). And the findings are fascinating: A trained version of the new WaveNet artificial neural network produces English and Chinese speech that sounds more natural than Google’s latest implementations of two kinds of TTS systems. […]
Share this Article
Recommended