I have an idea to adjust the text to speech function according to the character speaking, but I lack the power to influence the reader industry. My idea is to use a color coded colon to distinguish the characters. The main character in my novel could use a previously prescribed default to limit the markup. Here’s an example:

The main character says this and can possibly be left unmarked.

:The secondary character is marked by a color coded colon and theoretically a color coded period could act as the end mark to switch the voice back to the default.

I don’t need a separate narrator because my novel uses the first person perspective, so others would probably use a color coded colon to mark all characters. Reducing the markup is the Ideal strategy. The colon is used by playwrights, but I’ve found it best not to skip a space after the colon renders and the color code eliminates the use of the actual name. Characters still need an introduction such as, “What’s up, Harry,” but that’s no big deal.

TTS apps tend to voice out things such as underscore which is the reason I used the colon, but I’m sure alternative methods exist.