April 3, 2018 05:03 pm PDT
Original Link: http://feeds.boingboing.net/~r/boingboing/iBag/~3/8rUaw9FETIg/googles-talking-ai-is-indist.html
Google's talking AI is indistinguishable from humans
Tacotron 2 is Google's new text-to-speech system, and as heard in the samples below, it sounds indistinguishable from humans.
From Quartz:
The system is Googles second official generation of the technology, which consists of two deep neural networks. The first network translates the text into a spectrogram (pdf), a visual way to represent audio frequencies over time. That spectrogram is then fed into WaveNet, a system from Alphabets AI research lab DeepMind, which reads the chart and generates the corresponding audio elements accordingly.
Tacotron 2 or Human?
In the following examples, one is generated by Tacotron 2, and one is the recording of a human, but which is which?
Soundwave image by T-flex/Shutterstock.
Original Link: http://feeds.boingboing.net/~r/boingboing/iBag/~3/8rUaw9FETIg/googles-talking-ai-is-indist.html
Share this article:
Tweet
View Full Article