All these AI text to voice models seem to ignore emotion. It always sounds like a robot.
That should actually be possible to do already with existing tech. I haven't seen if you can instruct Kokoro to read in a certain way, does anyone know if this is possible?
https://emosphere-tts.github.io/
We are getting there
https://www.microsoft.com/en-us/research/project/emoctrl-tts...