There is no way I could write a book - ironically, the one thing I plan to do on my bucket list of one item! The elusive gap for me is the vast rift between telling a story and typing one out, while maintaining the same organics and start/end structure. Its surprising what people take for granted, as in the complexity of a one-on0one conversation. So what about a dynamic pair of friends that react to the pauses in your speech? The idea is/was to let a dynamic 3d person's reactions help you story tell more fluidly for dictation and transcription. Yes a phone app rolls out a set of changing expressions while it audio records (for later transcription).
How would your story change if your "friends" gave a confused look when you mention a specific detail? Ide bet the path and tone of your dialog would take a sharp turn!
Or this goofy-ass face... begging to hear more about what you just said, or encouraging the direction you are going,