[gepoch]

I've been teaching myself Gregg simplified since late last year. I probably put in 30 minutes a day, and an hour or two on the weekends.

First of all, it's very fun (the main reason I got into it)

I currently can write about as fast as I type. Somewhere in the range of 40 - 80 WPM depending on how many complex words I need to use (those slow you down a lot more than common ones)

The biggest negative for me right now: reading. I can write it far faster than I can read it. For things like meeting notes,this pretty much rules shorthand out as I need to be able to flip through to find a particular meeting, or something important that someone said. This will get a lot easier with practice, but it's the biggest blocker for me at the moment.

I've made about 1000 flash cards in Anki for memorizing words, short forms, special rules, etc.

Very cool, but probably an obsolete technology in an era where recording is easy and typing is pretty good. I might feel differently as I improve, but that's how I feel now.

[delinka]

I think the original intent was to take notes quickly and later transcribe them to be read as normal text. I've wanted to learn shorthand for writing into my tablet and have it transcribe automatically, but I haven't put in the research to know if a translation app exists.

[gepoch]

Right, that's definitely the intent of shorthand, but it works better when you have someone dedicated to processing the captured conversation.

If I tried to transcribe my notes from every meeting, I would end up working on it for more time than the actual meeting took at this point.

I have thought about the exact same app actually. Being able to photograph my Gregg outlines and get an English transcription would be the dream.

Gregg and many other shorthands are purely phonetic. There's no difference between to, too, and two. You would need to do OCR on the outlines, map those to sounds, and map the sounds to possible words, then assemble a meaningful sentence based on what meanings make sense in context. A neural net could probably do a good job of mapping from outlines to possible words, and you could use NLP tools to try to choose a sentence from the possible outputs. It would be tricky to do well, for sure.

For now, it's vim for my meeting notes :)

[0xdeadbeefbabe]

> research to know if a translation app exists

You need something easy for a human to write and easy for a computer to decode.

I think Gregg is out for this reason, but so is graffiti (it's not fast enough right?).

[finder83]

There's a site that I've been following on which the author regularly uploads transcriptions into various Gregg versions, including simplified -- https://gregg-shorthand.com/?tag=readingmaterial+simplified

I'm still early in learning, so I'm not sure how much it helps, but I've heard that regularly reading material written in Gregg that you don't know beforehand is the fastest way to increase your reading speed.