Inconsistency in volume of Ai Voices for CC

So the current project I am working on I am not noticing this as much, not sure if the 13 release included a fix? But i am noticing a better consistency with the volume, also playing around with where you put the breaks helps.

What I did do to address the problem in the other project was to go into the volume settings on the sound clip and either fiddle with the dB or selected Normalize under audio recording and that seemed to help. 

I did notice that the generated voice still has issues with pronunciations, especially of the word Ombudsman, which I use a LOT because of my line of work. lol. For some reason it will say it right, but other instances will say OH-budsman, I typically swap the capitol O out with lower case and re-run it and it fixes it, which is totally odd. lol.  It just doesn't like the word I guess.

I've experienced this as well. Sometimes it corrects itself if you generate the audio again, but I haven't found any pattern to it yet. Have you discovered anything new since you first posted this?

Yes, I’ve noticed that too.
This issue often arises when captions are split across multiple lines or timing segments.

I’m interested to hear how others handle this—any best practices or creative solutions?