Instead of having to re-render all the TTS at once, we want the ability to choose a single string to re-generate.
Instead of having to re-render all the TTS at once, we want the ability to choose a single string to re-generate.