This engine can be used as follows. 1. Record 2. Take the first word and figure out how long it takes for you to say it. 3. Put that in the () seconds part of the block. 4. Parse the word into IPA-ish. 5. Put that in the other part of the block. 6. Repeat 2-5 for the rest of the phrase, using waits where necessary.
I did it all. If you want it to be better, I need better art.