Today I tested the text to audio conversion part and I was disappointed. I used the same text on an image from the Star Trek Online game and then used a python script to combine the videos.
Grok is one of the most tested AIs on the animation side by me. I use it for content and testing and I can say that it is limited to converting text to audio, gymnastic moves with animals: jumping over your head... It is easy to be misled by animated combinations of words. I control the movement part between the character and the rendered camera mode quite well. What disappointed me the most is the audio part.
This is the text I used with Grok AI , bad sound result:
She is confident and introduces herself to the audience with these words: "Hello, Captain, I'm Beverly Crusher, born Beverly Howard on October 13, 2324, in Copernicus City, Luna. I attended Starfleet Academy from 2342 to 2350. Some people call me the Dancing Doctor."
