In the last few years, text-to-speech motors have produced huge progress thanks to contemporary machine learning methods. It was previously unbelievably simple to understand that it was a computer reading a text rather than a human being. But it’s rapidly evolving.
Today, the Amazon AWS cloud computing arm introduced a number of neural text-to-speech models and a fresh style of newscasters.
Amazon today announced the general availability of the Neural Text-To-Speech style and newscaster in Amazon Polly, it is cloud service that converts text to speech, which cannot be overstated by Google’s WaveNet, which simulates stresses and intonation by defining the tonal pattern.
As tech evangelist Julien Simon pointed out in a blog post from Amazon Web Services, neural text speech offers considerable quality improvements in the speech by increased naturalness and expression.
As the new service is formally referred to, Amazon Pole Newscaster is the consequence of years of text-to-speech studies that AWS is now also making accessible via its neural motor.
This fresh engine has 11 voices, three for the U.K, eight for the U.S. This is no difference to comparable neural motors like Google’s WaveNet and others.
In this era of false news, it is at first a bit difficult to have life-like robot voices which sound like true newscasters. However, it doesn’t make much difference whether a robot or a man reads the text. There are many excellent voice uses, and you can hear these voices much longer, considering the examples provided by AWS, than those of the ancient voices before you want to cut your hearing.