New Delhi: Alexa’s voice will be customised by Amazon! It appears that the business intends to enable Alexa, its virtual assistant, to impersonate any speech.
Amazon is developing a technique that will enable Alexa to imitate anyone’s voice after hearing less than a minute of any audio, according to Reuters. Rohit Prasad from Amazon revealed the information during a convention in Las Vegas.
What about your departed family members?
In light of the Covid-19 pandemic, in which “so many of us have lost someone we love,” Prasad claimed that the purpose of this new function would be to “make the memories last.”
The usage of such technology is contradictory even when done with the best of intentions. Microsoft terminated its AI capabilities that might be used to impersonate voices just two days ago. The concern? Deepfake clips might be made with it, particularly for politicians. Hackers might also use it to extort people.
Prasad claims that “generalisable intelligence” is Amazon’s goal for Alexa. Amazon essentially wants Alexa to be able to adapt to all user contexts and pick up new ideas on her own. The all-knowing, all-capable, super artificial general intelligence, he continued, should not be confused with this.
Amazon demonstrated how Alexa, a feature of its Echo devices and Android smartphones, could be used to impersonate anyone’s voice during the conference. A young child requested Alexa, “Alexa, can grandma finish reading me The Wizard of Oz?” Not long after, Alexa’s voice changed, sounding more like the child’s grandmother and less robotic.
