Happy to say that as we get closer to a Early Access release, Synthasia now has builtin Text To Speech !
We are using webgpu to bring a hardware accelerated voice generation experience across multiple platforms, meaning that out of the box, without further configuration from users (at least in most cases... i hope) , you will have the game narrated to you by a cast of multiple AI Voices, with each npc having a different voice (well, up to a certain extent, the voice pool is limited).
Made a very fast video showcasing a little bit of it.