Alexa will soon be able to mimic the voice of your relatives — alive or dead.
At the re:MARS conference, Amazon’s global artificial-intelligence event in Las Vegas on Wednesday, Amazon unveiled various new technologies, including the new voice mimicking ability of its virtual assistant Alexa.
Presenter Rohit Prasad, the senior vice president and head scientist for Alexa, explained that human-like attributes, such as empathy, are key for building trust.
“These attributes have become even more important during the ongoing pandemic when so many of us have lost ones that we love,” Prasad said. “While AI can’t eliminate that pain of loss, it can definitely make their memories last.”
In a video demo, the camera zooms in on an Alexa while a child asks the AI device: “Alexa, can grandma finish reading me the Wizard of Oz?” to which the system replies, “OK,” and starts reading in the child’s grandmother’s voice.
Prasad explained that they had to learn how to create this high quality voice filter with less than a minute of an audio sample. According to The Associated Press, Amazon didn’t provide further details about the feature.
Amazon already has settings that allow Alexa to replicate the voices of former NBA player Shaquille O’Neal and actor Melissa McCarthy, although this process had required hours of audio recordings, according to Variety.
“We are unquestionably living in the golden era of AI, where our dreams and science fiction are becoming a reality,” Prasad said, adding that this advancement would “enrich daily lives of everyone, everywhere.”
This news comes on the heels of Microsoft publishing a set of new AI rules, which would set limits on who would be able to create it and how it would be used.
“It is … easy to imagine how it could be used to inappropriately impersonate speakers and deceive listeners,” said Natasha Crampton, Microsoft’s chief responsible AI officer, per The Guardian.