My Morning MP3

I have a rough idea for a product I want to build for myself, and I already have several of the pieces of technology to make it work. But I’m not sure exactly what to build, so I’m going to think out loud while I work it out.

The goal is to have an MP3 of something to listen to in the morning. Why? Mostly because I think it’d be cool, which is a terribly reason to work on something. Secondarily because I want to learn the skills for doing it, which is fine for teaching yourself, but the result isn’t necessarily something anyone would actually need or use. So I should purely in the mindset of designing a thing for one person, me. To see what happens.

So if I were to wake up in the morning and load up an MP3 while I walked my dog, what would I want to hear?

One idea is to make some sort of chatGPT radio show. I could imagine a series of segments, some topical and some not. It could connect to my calendar to see what I have on that day. I could see connecting to my todo list. Scraping the latest news stories. Maybe some things could be more creative, like little two minute stories. Maybe it could pull down some new music from somewhere.

There are summarisation APIs now, where I can point to anything — Podcast, YouTube clip, webpage — and it will summarise it for me. I could imagine pulling from some interesting sources to summarise things while I walk. That could be interesting.

But maybe I’m barking up the wrong tree. Maybe these are a few tech demos.

  • I want to learn how to translate words into an MP3
  • I want to learn how to summarise with Kagi’s universal summariser
  • I want to learn how to get chatGPT to talk to an external source

So I don’t think I have a product yet, even for an audience of one. I have some R&D I want to tinker with.