Amazon Developing Newscaster-Style Speaking Voice for Alexa
Looks like Amazon’s Alexa voice assistant will soon be reading the news articles for you just like a professional newscaster. According to a report by The Verge, Amazon is developing a bunch of new speaking styles for Alexa, including a ‘newscaster’ voice for reading articles, that will be rolled out on compatible devices in the next few weeks.
Samples of some of the Alexa voices being developed by Amazon are embedded in the end, all of which are quite impressive, thanks to the company’s development of “neural text-to-speech” technology or NTTS.
The newscaster voice particularly seems to incorporates stresses into sentences in the same way you’d expect from a TV or radio newscaster. Amazon says its survey has found that the majority of users prefer it to Alexa’s regular speaking style when listening to news stories.
“This is the next generation of speech synthesis, that use machine learning to generate expressive voices more quickly. Currently, Alexa uses concatenative speech synthesis, a method that’s been around for decades. This involves breaking up speech samples into distinct sounds (known as phonemes) and then stitching them back together to form new words and sentences.”
Amazon says the newscaster speaking voice was created by recording audio clips from real life news channels and then using machine learning to spot patterns in how newscasters read the text.
Check out the samples below to get a better idea of how the different voices compare.
Neural text-to-speech (NTTS):