Sam Witteveen and I were invited over to Google's Mountain View location (in Silicon Valley) for the Google Developer Expert DevFest at the beginning of November. As luck would have it, our friend Vikram Tiwari organises the San Francisco Google Developer Group Cloud MeetUp, and invited us to take part in an event in San Francisco : Experts Panel - Voice and Machine learning.
As part of the discussion, the panelists were each given the opportunity to do a short presentation, and for my part, I gave a talk titled "Deep Learning Voices", which discussed the history of different approaches for the 'features to audio' stage of the processing pipeline. Obviously, WaveNet (in its various forms) made an appearance, but the talk also extended to WaveRNN and the GlowNet / FloWaveNet approaches that recently came out of Nvidia and Korea respectively.
The slides for my talk are here :
If there are any questions about the presentation please ask below, or contact me using the details given on the slides themselves.