I've been using Googles studio and this is much better. I love it. The clone worked fantastic.

I disagree with everyone who says they should not charge credits for previews. Everyone has to figure out how to use the tool as economically as possible and allowing people to squander credits freely, and carelessly will just spoil it for the rest of us.

If this is a concern then turn off the auto generate switch and manually generate speech only when you are ready to test your changes. When I first started with the tool, I was tweaking everything back and forth with the entire project selected which triggered automatic regeneration of all 1300 blocks. It took hours to do and consumed a lot of credits. Now I learned how to not squander those resources and I'm thrilled about it. My biggest concern is that if they make the tweaking and testing free, no credits, then people will do what I did and squander them an put them out of business. Then we all lose.

Alternatively, if customers pressure them hard enough, we might force them to disable the costly features that squander so much resources, which would be bad for everyone who is willing to learn how to plan our testing economically. If you need more testing credits then get a bigger plan.

Ellis Nguyen has been very responsive, solving all my issues quickly. My only concern is that there still seems to be a lot of issues to resolve, which is just growing pains I'm sure. As long as he continues to solve my issues as they arrive, then I'm a thrilled customer.

Ellis is aware of my feature requests already, which include:
1) better control of organizing sections. It seems to automatically chunk 50 blocks to each section but I'd like to organize them by chapters. If a chapter is too big to fit in a section then I don't mind having multiple section per chapter but I want better control on organization of the content.
2) I'd like to be able to do partial exports of one or more sections. I don't know how to do any kind of partial export yet, if that's even possible, but I'd like to be able to have more control over chunks of larger content.
3) The interpreter has to be better at figuring out how to pronounce symbols like = and %.
4) The AI needs to be able to figure out if an apostrophe or a quote is for text string or to imply the foot in a dimension, such as a 2x4.
5) The AI needs to be able to figure out if the x in 2x4 is an x or a times or a "by". Same for "/".
6) I haven't figured out how to remove the dead air, pauses in the text, or to insert controlled pauses. I thought a tab after a number would suggest a pause but it doesn't. It reads the bullet point text after the number as if it's one word without a tab or even a space and it sometimes inserts pauses where there isn't any reason to do so. Not a big deal at all but just something to monitor.

I'm really loving this tool though. It's the best I've ever seen, and I've seen a lot. I take a lot of courses and I like to listen to the textbooks while I drive. Using my own voice adds a significant level of intrigue. I seem to pay more attention to the content, as if I'm trying to recall what I was thinking when I read that even though I know I didn't read it.