Reading at 3x
A couple of years ago I was introduced to the term "immersive reading," which means listening to the audiobook while you read along in the physical book. I grew up reading physical books and still love them, and I've spent a lot of time listening to audiobooks on road trips, so this felt like the perfect mix of the two.
The more I got into it, the more I realized it's my favorite way to read. It keeps me focused and really locks me into the story or topic, and in an era where attention spans are clearly getting shorter, I'd recommend it to anyone. It also helps you get through books faster. When I first started I listened at around 1.5x, but now I'm consistently at around 3x. That may sound ridiculous, but I think what's happening is something called reading ahead, where the audio plays in the background while my eyes run slightly ahead of it, and my brain processes the two on separate threads. It's hard to explain, but it has a very distinct feel, and I'd encourage anyone to try it.
The problem with articles
This has created a bit of a problem, though, because any article without an audio option is now something I struggle to get through. It also turned out to be an opportunity. I've used ElevenLabs for a lot of fun audio projects, but it's quite expensive, and its interface is built for professional workflows, which makes it more complicated than I need when I just want a quick audio version of an article. So I'm working on a small side project called Article Tapes that creates audio for any article I'm reading, and you can either run a local model on your own computer or bring your own provider.
Reading my own blog
Audio also got me thinking about this blog. If I'm going to be such a proponent of immersive reading, my own posts should probably have a native audio option. I could have used a model for this, but reading the posts myself felt like a better way to show my personality and make the blog feel more human. The problem was that whenever I tried to record a full take, I'd make a bunch of mistakes and end up throwing it out.
In the era of custom software, though, that's a solvable problem. I had Claude Code build me a kind of recording studio that lets me record a post in chunks and then stitches them together automatically. That's how the audio for this post was made, across many takes rather than one.
The other cool part is that I can edit my writing while I record. If a sentence feels unnatural when I say it out loud, which happened a decent amount, I just change it on the spot, and when the program stitches the final recording together it applies those edits back to the post automatically.
All that to say, it's a really fun time to be an immersive reader, and I'd encourage everyone to try it out and get creative with it.