Digital assistants like Amazon’s Alexa come in a variety of voices and accents that users can choose from, but what if you wanted Alexa to sound like someone you know? It turns out that is something that Amazon is working on, as reported by TechCrunch during Amazon’s re:Mars conference that was held in Las Vegas today.
With this upcoming feature, it seems that just based on one minute of speech, Alexa will be able to mimic the voice of anyone, alive or dead. This means that if you have a loved one that has passed that you really miss, if you could feed Alexa their voice from a recording, the digital assistant could sound like them. According to Amazon’s Senior Vice President and Head Scientist for Alexa, Rohit Prasad:
“This required inventions where we had to learn to produce a high-quality voice with less than a minute of recording versus hours of recording in the studio. The way we made it happen is by framing the problem as a voice conversion task and not a speech generation path. We are unquestionably living in the golden era of AI, where our dreams and science fictions are becoming a reality.”
There are a lot of implications behind this tech and while it does sound noble in its intentions, understandably some might be wary of how this could be used for deepfake scenarios. Amazon did not mention when the feature is expected to roll out, but it sounds intriguing from a tech point of view.