Google’s DeepMind AI Creates a Machine That Sounds Exactly Like a Human




Google has announced a new voice synthesis program in WaveNet, powered by deep neural AI. Understanding voice samples has been powering programs like Google Voice Search for quite some time now. However, synthesizing something from those samples is proving to be a challenge. Now, Instead of simply analyzing the audio it's fed, it learns from it.

By working with at least 16,000 samples per second, WaveNet can generate its own raw audio samples with minimal human intervention; it uses statistics to actually predict which audio piece it needs, what it has to 'say' next.





<script language="javascript" type="text/javascript" src=";pubUrl=[SRC_URL_ENCODED]&amp;nofr=1"></script>


"The Galaxy" in Your Inbox, Free, Daily