I think that's impossible to make, if you meant it pronounces correctly, and really accurate, the sounds file is going to be alot... Except if you record the each word sound, or use the AI reading the text.
As for lip syncing, it can be as simple as mouth opening and close.