By: Ramon Martinez
Have you ever wanted Alexa to sound like Walter Cronkite or, for those of you too young to know who that is, Anderson Cooper? Well, you might not get exactly that, but Amazon is developing new speaking voices for their resident voice assistant, including that of a newscaster.
The newscaster voice would be available to read your news articles, making it sound more “newsy.” That voice will launch before the end of the year on all Alexa-enabled devices. Below you can hear samples of what it would sound like. The delivery is more assertive and punches different parts of a sentence, much the way a newscaster seems to be trained to do.
A blog post on Amazon explains how this works. “When people speak, they use different speaking styles depending on context. A TV newscaster, for example, will use a very different style when conveying the day’s headlines than a parent will when reading a bedtime story,” the blog post reads. “Amazon scientists have shown that our latest text-to-speech (TTS) system, which uses a generative neural network, can learn to employ a newscaster style from just a few hours of training.”
It goes onto explain how synthetic speech through the TTS system sounds much more natural because it’s not deriving each of its sounds or words from a database and stringing them together like common concatenative methods. NTTS (neural text-to-speech) allows Amazon to create a more natural speech, replete with a lot of the nuances of normal everyday sentences.
Machine learning was used to analyze recordings of actual newscasters delivering the news and break down what the nuances were that made up the speech pattern. Apparently, this only took a few hours of analyzed data to result in this now crystal-clear newscaster cadence. You’d expect it would take hundreds if not thousands of hours for it to learn this pattern, but no. That’s AI, folks. It’s smarter than humans and faster too. The last words before the whimper.
The new Alexa voice will be available in both male and female. And considering the speed with which this voice was created, we can almost definitely expect a plethora of new voices in the next year.
ABOUT THE AUTHOR
