It said on its official blog that with the new machine learning approach, NTTS will deliver “significant improvements in speech quality.” The 2 things it increases are naturalness & expressiveness. Both, said Amazon, were key to synthesizing lifelike speech that is getting closer than ever from human voices.
As of today, NTTS will be available for 11 voices, both in real-time & in batch mode:
- All 3 UK English voices
- All 8 US English voices
Introducing the newscaster style
Speech quality is certainly important, but more can be done to make a synthetic voice sound even more realistic & engaging. What about style?
For sure, human ears can tell the difference between a newscast, a sportscast, a university class & so on; indeed, most humans adopt the right style of speech for the right context, & this certainly helps in getting their message across.
With this new feature, from news to blog posts, narration will sound even more realistic, said Amazon.
Click here to read the post.