Amazon has displayed a glimpse into Alexa's newest feature, the assistant will mimic the voice of family members- of the ones who are gone too. This announcement was made at Re: Mars conference in Las Vegas. The specialty would be Alexa would only need less than a minute of that person's recorded voice to replicate the same.
According to Rohit Prasad, senior head scientist for Alexa and vice chairman for Alexa at Amazon, that the will behind the feature was to build greater trust in the interactions users have with Alexa by investing more "human attributes of empathy and affect"."These attributes became even more important during such times of pandemic where we have lost so many lives, lives of our loved ones" Prasad said. "While AI can't eliminate that pain of loss, it can definitely make their memories last."
In a demo video played at the event, a kid asks "Alexa, can Grandma finish reading me the 'Wizard of Oz'?" Alexa then processes the request and switches to another voice replicating the child's grandmother. Alexa then reads the book in his grandmother's voice.
To create the feature, Rohit said the company had to learn how to make a "high-quality voice" with a shorter voice sample, as opposed to hours of recording in a studio.