Us there any AI software that can do this?

[DaneClark]

Are there any ai programs where I can make muffled voices sound clearer by typing in the words being said and then let the ai clone and “trace over” the muffled voices while recreating the vowel and consonant sounds based on the words I typed in? I know that there are lots of ai voice cloning tools out there, and I know that some if them that let you trace cloned voices over clips of people talking, but I need something that also uses some kind of text-to-soeech system to clean up the cloned voices that are being traced.

I am trying to restore something for the Lost Media community that only exists with poor audio quality and being able to do specifically that feels like it could be the only way to restore it into something presentable and it really feels like it should be theoretically possible to do that.

[davexnet]

I don't know of such a thing. I assume the audio is at least intelligible.
Can you post a sample, perhaps a few minutes?

[DaneClark]

I don't know of such a thing. I assume the audio is at least intelligible.
Can you post a sample, perhaps a few minutes?

There are only a few lines where I can't make out what is being said, I'll pm you an audio file of the whole thing (it's 10 minutes) along with a transcript I made when I have time.