RevEng Posted yesterday at 05:08 PM Share Posted yesterday at 05:08 PM I've been playing around with Google's NotebookLM product. It's an AI tool intended to help you extract information to better understand the material contained in a set of documents. (could be research papers, school texts, audio lectures, etc.) It can generate information summaries, FAQs, Table of Contents, Timelines, you can interrogate the AI about the info in the documents, and... it can generate an AI two person podcast. Naturally, I generated a podcast from the 7800 Software Guide. I think the description of holey DMA is lacking the right context, and they mispronounce "Maria" several times... but otherwise it's fun and pretty engaging! Check it out at the MARIA Documents section of the 7800 wiki! 5 1 Quote Link to comment Share on other sites More sharing options...
Eagle Posted yesterday at 05:17 PM Share Posted yesterday at 05:17 PM 😂😂😂 Fantastic 2 Quote Link to comment Share on other sites More sharing options...
+Muddyfunster Posted yesterday at 05:27 PM Share Posted yesterday at 05:27 PM When I heard "Mariah", I got very scared. It's almost time to defrost her after all! Thanks for linking and sharing Mike 2 Quote Link to comment Share on other sites More sharing options...
+Karl G Posted yesterday at 05:37 PM Share Posted yesterday at 05:37 PM It's really quite remarkable how much it gets right, even if there were three different pronunciations for "Maria". 🙂 2 Quote Link to comment Share on other sites More sharing options...
RevEng Posted yesterday at 05:46 PM Author Share Posted yesterday at 05:46 PM Lol. Yeah, the Maria pronunciation really gets on my nerves, mainly because the content is pretty good otherwise. I tried creating another document that contained the proper pronunciation of "Maria", and the notebook interface correctly identified the doc as containing pronunciation information for "maria", but it didn't help. I think the script generation AI is a bit uncoupled from the voice generation AI. According to this youtube interview, they're looking at adding stuff to guide the podcast content. Hopefully they add in something to fix the pronunciation too. 1 Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.