We have a lot of tapes around of his voice, from voice mails to family videos to some things from his work. If you are open to reaching out that would be awesome, I’ll check out the site as well.
Edit: I’ve wanted to make some sort of soundboard + “text to talk” setup for this family member. He often can’t participate in conversations because he writes on a whiteboard, and the speed of chatter moves faster than his writing
We also have an API that you might find useful for the soundboard project: https://app.resemble.ai/docs
Out of interest what are the average response times to generate a clip of one or two sentences from a configured voice?
Imagining the easy text-to-speech solution the OP could build on this resemble API.