Replacing espeak with the state-of-the-art TTSs?

[ebraminio]

espeakbox is just complete and fantastic, period.

I was thinking about some perhaps more updated tts models like this (dockerhub) or Baidu's (on GitHub, 2, 1) or this to be provided on such easy to use format like this project which first needs evaluating them and see if they fit for such work.