Quick Facts: > Needs a decent set of audio data to work with, 30 minutes of audio book will do > They are working on watermarking the waveform some how try and

Quick Facts: > Needs a decent set of audio data to work with, 30 minutes of audio book will do > They are working on watermarking the waveform some how try and hamper it being used to impersonate people in a illegal way. Obviously bringing this tech to the masses is an interesting development. In a post Donald Trump world, anything we can do to get people to use critical thinking is great, perhaps eventually we will all develop a photoshop radar for audio too. Its still pretty crazy to me what we will be able to produce with audio and visual manipulations on the fly, as well as advanced AI.