READ IN: Español

Using Artificial Intelligence techniques, Amazon is working on enabling its users to recreate anyone’s voice with Alexa.

Recreating anyone’s voice could be possible by synthesising short clips into longer clips. And that’s exactly what Amazon is trying to do through the Alexa virtual assistant. During the company’s annual re:MARS event, a sample of this project was performed, using the voice of a deceased grandmother reading a story to her grandson.

“This required inventions where we had to learn how to produce a high quality voice with less than a minute of recording versus hours of recording in the studio, the way we did it is by framing the problem as a voice conversion task and not a voice generation path,” said Rohit Prasad, Senior Vice President and Chief Scientist of Alexa at Amazon. “We are certainly living in the golden age of AI, where our scientific dreams and fictions are coming true.”