Amazon is developing a technology based on Artificial Intelligence (AI) that recovers the voice of deceased loved ones and reproduces them through its virtual assistant, Alexa.
The company revealed this work during the re:MARS conference, an event dedicated to machine learning (machine learning) and automation, where it exhibited a demonstration of it.
The person in charge of making it known was the senior engineer Rohit Prasad, who assured that “one of the most surprising things about Alexa is the relationship of companionship that we have with her.”
Prasad insisted that “in this relationship of fellowship, the human attributes of empathy and affection are key to building trust” and that these have become relevant as a result of the pandemic.
“Many of us have lost a loved one, and while AI can’t take away the pain of loss, it can make their memories live on,” Prasad said.
So, the expert exposed a case in which a child asks Alexa if her grandmother can finish reading “The Wizard of Oz” to him. After receiving approval from the assistant, you can listen to the narration of a part of this book with a different female voice.
The manager stressed that this invention required Amazon developers to learn to “produce a high-quality voice with less than a minute of recording, versus hours of recording in the studio.” And he has anticipated that, in order to work on it, they have focused on this skill as a voice conversion task instead of generating that voice.
K. Tovar
Click the link to subscribe to our news and media group on Telegram: https://t.me/G_ELSUMARIO_News
Source: LaTercera
