I stumbled upon the perfect reason to use Gemini and NotebookLM on a gaming handheld

Android Arcade

With Android Arcade, Android Central Senior Editor Android Myrick takes a weekly deep dive into retro gaming, emulation, handhelds, and everything in between. Do you remember the days of needing to go to the store to pick up a game guide because you couldn’t figure out how to beat it? Yeah, that still happens to me, except I don’t have to go to the store, as the answer is just a Google search away. But I’m not always playing games on my computer, and even if I’m playing a handheld connected to the internet, I don’t want to sift through a bunch of links that may or may not have what I’m looking for.

Last month, the Anbernic RG557 was released, and a recent update introduced “Anbernic AI.” Essentially, it’s a chatbot that resides on your gaming handheld, designed to provide answers and even translations. After a bit of digging, I found that it’s reliant on either Deepseek or Alibaba’s Qwen LLMs. That doesn’t leave me feeling great, but it got me wondering if I could just make one myself.

Gemini Gems and ChatGPT GPTs, I set out trying to replicate what Anbernic did, but using AI chatbots that I trust a bit more and trying to get them working on the handheld. It didn’t quite work out how I had hoped, but fortunately, that’s where NotebookLM came to my rescue.

NotebookLM to the rescue

Then came along another tangentially related idea. Google has been super-charging NotebookLM, but it’s one of those tools that I always seem to forget about. There are a few projects that I have in which NotebookLM could be useful for me, but first, I was curious about how it would do as a gaming encyclopedia of sorts.

There are limitations with NotebookLM, as you can upload only 50 “sources” to one notebook. However, thanks to being subscribed to Google One AI Pro, I’m able to add 300 sources. I say “only” because the idea is to source and upload strategy guides for a variety of retro games.

Needless to say, there are many more than 300 out there, so I opted to take a different approach. Instead, I decided to create notebooks based on the series of games, meaning I’ll have one for Mario, Zelda, Final Fantasy, Fallout, and so on.

Thanks to the gaming community, a lot of those old Nintendo Power and Prima strategy guides have been archived. It was simply a matter of sourcing, downloading, and uploading them all to NotebookLM.

However, there was the small problem that many of these old strategy guides are stored in either a CBR or CBZ format. This is no good, as NotebookLM only recognizes PDF, TXT, Markdown, or Audio files when uploading. On the bright side, some quick research revealed that you can convert those CBR or CBZ files to PDF using Calibre. And it’s much faster than trying to use any of the “free” online conversion tools.

With that out of the way, I started uploading a few files and then just waited for that to finish. You’d be surprised at how large some of these PDF files can be, and NotebookLM has a size limit of 200MB per file, so keep that in mind.

A little while later, I had a one-stop shop for resources pertaining to a few of the Super Mario games that I’ve played over the years. Now, whenever I can’t remember how to get past Bowser, I can just pull up NotebookLM and ask.

I know that this took a bit of a left turn, considering my original goal, but the overall point is to demonstrate that AI tools can be useful in ways you might not have considered. I don’t even know where the idea to create these guides came from, but I’m sure glad it came to me.

No, you aren’t able to use it while you’re playing like you can with Gemini or ChatGPT. But I have to wonder whether that functionality will be implemented in the near future. I’d really love it if there were a way to create a Gem or GPT that could be used with Gemini Live, but used NotebookLM as the source.

Nevertheless, who knows what these companies are going to bring next? Maybe something else will come along that takes all of this to the next level, and we just don’t know it yet.

BetaBeacon
I stumbled upon the perfect reason to use Gemini and NotebookLM on a gaming handheld