Skip to main content
Synthesizing realistic speech is difficult largely because of the one-to-many problem in linguistics: one text string corresponds to infinite possible acoustic realizations. Strings of words can also be pronounced differently for special effect. Rime is opinionated about the default acoustic realization, and the voices speak fluently and correctly out of the box. You can still customize the output. The API also allows you to make extremely low-level adjustments for your particular use case, such as custom pauses and custom pronunciations. To see all the available customization options, check the pages in the Customizing Mist group in the menu.