[Wald76]

VoiceXML uses SSML to mark up text to be spoken to the user, and in fact the two standards were developed in parallel by the W3C’s Voice Browser Working Group.

[jquip]

Hopefully, with an AWS offering, this could give an impetus to widespread adoption. A lot of great ideas are simply left dormant because of network effects.

[dragonwriter]

SSML is fairly widely supported now in it's domain (and, unless I'm mistaken, the generally similar features in CSS Speech were modelled on it.)