I Use AI to Generate My Voice (it’s eerie)

The audio above is an AI-generated audio file of text I’ve typed after having provided a couple sentences of recorded speech. It’s eerily accurate, of course. This is the future, after all.

I’m thinking of something I’m sure has been thought of by others, but I haven’t heard it discussed in this particular way. I mean, we’ve heard about sometime way off in the future being able to store our mind somewhere and have it live on beyond our natural existence.

But I think that time is now.

I visit with my elderly parents almost every weekend and I’ve been recording our conversations for several years.

As I explore AI more, it’s occurred to me that it could be used to generate experiences in the present with loved ones who have passed on. But would it be a positive thing?

I envision feeding a language model the text from all of our conversations. I envision feeding another model my mother and father’s voices. And then I envision continuing to have conversations with them.

I imagine it would be much like chatting with Jasper (an online ai tool). Some replies would be accurate and some would merely be conversational; not entirely accurate in terms of past events. How is that really any different than now? When they are alive?

I imagine it would allow future generations to ask “them” questions and get pertinent replies. You might program a script to call grandkids on their birthdays.

Video is coming soon, so I imagine you might add AI generated video and be able to have video conversations with your grandparents at any time.

It’s possible today to load a PDF document into a language model and then chat with it and it will respond based on what is in the PDF document. I imagine you could scrape enough content scattered through facebook posts, emails, letters, and videos of your long lost loved one and very closely approximate having them stored in cloud.

Is that a positive thing?