This is actually pretty easy to set up and you can crank these out easily when you do.
This is also a speech-to-speech model (the Freelancer VA one) that I'm using here, which is fed Microsoft Edge TTS (yeah, really). You can feed any voice into it, however, once the model is trained, including your own.
I'll write down a tutorial on how to do this end to end so it should be within reach to anyone with an Nvidia GPU.
Meanwhile, I generated all the non-vanilla voicelines in Pennsylvania which you can download here, if any of the devs would like to try to make the audio work in game.