This is why I find using speech-to-text tools quite difficult to use: because the parts of my brain that I use for writing and the parts of my brain I use for speaking are a little different — although with significant overlap.
With writing I find I'm drafting the flow for readability and clarity as I'm writing, so I go back and rework bits and pieces — sometimes even while I'm in the middle to typing a sentence. Maybe it's because I write code for a living.
Speech only moves forward and you have to state your retractions or clarifications on the go. You can't go back and edit what you've said.
I've been trying to use speech-to-text a bit to: a) give my hands a bit of a break when I'm writing prose, and b) see if it's faster than typing.
I find there are long pauses while I'm struggling to draft what I'm going to say to what I want written, so I'm not sure if it is faster (given that I'm a ten finger touch typist so can type pretty fast is short bursts, and the time spent going back and tidying up the output which is somewhat tedious). It might improve with more practice.
— No tokens were harmed in the production of this comment. —
It's possible you have a touch of prosopagnosia, also known as face blindness. It would make sense as you would recognise people you interact with more regularly by other things (e.g.: voice, or even body or posture as the grandparent comment mentioned), but unfamiliar people tend to me harder to identify.
Thank you for the welcome and reminder. It made me laugh. It's a weird experience to have strangers analysing your eriting and thoughts. Interesting but weird.
I still clearly remember in my early twenties being stunned to discover that the Astérix comics are originally written in French and then translated. Coming up with names like Getafix in English for the druid — incroyable.
First personal little thing was a little mobile friendly website for taking coffee and tea orders. Two pages, one for someone to enter orders, second for listing orders and marking them as done. Was a bit of a gimmick idea I'd had in the back of my head for family lunches to save me running around with a scrap of paper and a pen asking what everyone wanted. Thought it would make a good little exploration of "vibe coding".
Second is a utility that will take a text file export from Wallabag, and use text to speech to convert it to an MP3 file. I then integrated it into a utility that I already wrote for managing articles so if I tag an article with listen it automatically gets converted and gets shunted to the podcast listening app on my phone.
Last was to recreate a little directory listing utility that I've had a 32-bit binary of for ages, but no longer (if ever) have the source code for. I'd always promised myself I would write it once I learned Rust, but decided try using AI as I was getting impatient to have it now. The utility lists all files, including hidden, in a directory, grouped by type. Below is an example listing the directory for the project itself:
reply