I still think there’s potentially a space for these dedicated AI-based voice recorders, but none of them are quite right yet. From David Pierce’s review of the NotePin:
In my time testing the NotePin, I’ve mostly had it around my neck, and I’ve used it to note reminders while driving, ramble long ideas to myself while walking the dog, and summarize calls and conversations. It’s certainly handy being able to just reach down, press the NotePin until it vibrates to indicate it’s recording, and then yammer away at nothing while my necklace dutifully listens.
Even though it works as advertised, David doesn’t think these need to be dedicated devices. iOS 18 and macOS Sequoia will have good transcription built in. The price at $169 also seems a bit high to me.
I never use Siri unless I’m in the car, and there I have exactly three uses:
- call someone
- respond to a text
- record a quick memo
That’s it. On road trips especially, my mind will wander, and I’ll have an idea for a new Micro.blog feature, or a blog post, or an edit I should make to my book. I don’t bother asking Siri for anything complicated because it’s not going to get it right.
For transcription, the current Siri isn’t very good. I expect it to get much better with Apple’s LLMs, but will it be as good as a more powerful model backed by OpenAI? And if not, is it worth buying a special device to get that extra quality?
My ideal AI device would be about the size of the NotePin or the Friend. Transcription would be effortless, with good sync of notes to other platforms or Dropbox. I could also ask it world knowledge questions like I do with ChatGPT. And as a bonus, sure, being able to talk to it and get comforting reassurance like the Friend would be neat too.
It doesn’t need a screen. It doesn’t need a camera. It doesn’t need a laser. It just needs to do these couple LLM-powered tasks exceptionally well.
