I already use TTS to listen to e-pulp fiction books, which is something a lot of people wouldn't put up with, but I could easily listen to books narrated by any

I already use TTS to listen to e-pulp fiction books, which is something a lot of people wouldn't put up with, but I could easily listen to books narrated by any of those sample voices, especially if it switched between distinct styles for each character without changing the voice. But that would still probably require human work to tag the dialogue with the right characters.