Any lifelike text-to-speech AI that is customizable in: whispering, pauses, slowing down?

I would like to leverage AI as text-to-speech. I don't need much accents, lifelike US/UK would be enough. The thing I'm looking for is extensible customization. I would like the voice to slow down in certain moments, whisper, make pauses, extend words.

The main goal is using it for relaxation techniques so it's pretty much necessary. Which provider or providers should I focus on?

accents artificial customization moments speech speech ai text text-to-speech voice whisper words