I have never seen an AI text to voice engine that I liked as it is so hard to be expressive with them. it's ok for straight forward text but not if it is too complex. I ended up getting some help from a friend which was great but I ran out of time and had to compromise on a lot of my original plans. TBH the audio was slapped together far too quickly over scenes that all still needed lots more love and attention... but hey this is just for fun and as a learning exercise so in the end I was fairly happy with it all.