The Cellar  

Go Back   The Cellar > Main > Technology

Technology Computing, programming, science, electronics, telecommunications, etc.

Reply
 
Thread Tools Display Modes
Old 07-21-2015, 05:36 AM   #1
it
Banned
 
Join Date: Dec 2011
Posts: 772
text recognition software

Following the conversation in the maps thread about handwriting, I was thinking of getting rid of mine.

Over the years I've accumulated a few heavy weight boxes of papers - mostly consisting of brain dumps. Some of them got lost through the journey but enough are still around to be a PITA each time I move, which is relatively often. Not to mention pretty difficult to navigate & organize, which is a big hamper on the already slim to non existent chance I will ever try to actually do anything with them... But I don't quite have the heart to throw them all out.

Short story shorter, I am looking for a good text recognition software that is able to recognize and save text from scans or - better yet - pictures.

It needs to be able to do pretty well with very tough writing. Imagine if you will a foreigner's english, with the hand eye coordination of a barbarian brute, the brain of a dyslexic child, and the reincarnated soul of the douche-bag who wrote the Voynich manuscript for the exact purpose of giving archaeologists a headache, and you'd pretty much be on the mark. Extra points if it's fluent in English and Hebrew.

Any suggestions?
it is offline   Reply With Quote
Old 07-21-2015, 10:32 AM   #2
footfootfoot
To shreds, you say?
 
Join Date: Aug 2004
Location: in the house and on the street-how many, many feet we meet!
Posts: 18,449
Hire a typist?

Doctors used to and may still do use typists to transcribe voice memos and other written notes.

I used to know an elderly, retired woman who could type like a fiend and would take anything and make it into a properly formatted word document. I don't remember what she charged but it had to be cheaper than scanning, converting, correcting and saving all that writing.

I've used several OCR applications over the years and none of them were very good, even with printed text, and more so if it was uneditable text, e.g. a jpeg of text.

Maybe things have changed but i don't tknow.

There is some sort of software that learns your hand writing and creates a typeface that looks just like your handwriting. That may be a step to then teaching an OCR application to recognize THAT typeface.

Good luck, keep us posted. I'd like to be able to do this because I write freehand and then have to type what I wrote, if I could scan it it might be easier since I am a mediocre typist at best, and can't afford to pay someone to type for me.
__________________
The internet is a hateful stew of vomit you can never take completely seriously. - Her Fobs
footfootfoot is offline   Reply With Quote
Old 07-21-2015, 01:05 PM   #3
limey
Encroaching on your decrees
 
Join Date: Feb 2004
Location: An island within the south-west coast of Scotland
Posts: 7,016
I use OCR software regularly to turn a text supplied to me as an image or pdf into an editable document as the first stage of my translation process. The software has improved dramatically in recent years with regard to printed text, being far more accurate than it used to be and able to deal with, for example, a printed document copied at a slant, or to reproduce tables. What it fails at is handwriting - there are too many variables.
I think footsie's idea of a typist is your best bet, to be honest. Checking and correcting a document that has been poorly picked up by OCR is teh debbil.
Maybe what you want is speech recognition software?? You read your texts aloud and the software produces a typed text. I believe that speech recognition software is pretty good at learning your little ways ...
__________________
Living it up on the edge ... of civilisation, within the southwest coast of

Last edited by limey; 07-21-2015 at 01:07 PM. Reason: I had an idea!
limey is offline   Reply With Quote
Old 07-21-2015, 01:12 PM   #4
glatt
 
Join Date: Jul 2003
Location: Arlington, VA
Posts: 27,717
Or if you have the time, reading the text aloud into your phone and let the phone convert to text. A few heavy boxes can't be much more than 5,000 pages, or about 500 - 1,000 hours of slowly reading it to your phone (assuming each page is text dense.) If it were me, I'd try reading a few pages into my phone and see how it goes and multiply it out to see how long it would take.
glatt is offline   Reply With Quote
Old 07-21-2015, 03:42 PM   #5
xoxoxoBruce
The future is unwritten
 
Join Date: Oct 2002
Posts: 71,105
My HP flatbed scanner translates printed text to a MS WORD document, but I doubt it would do handwriting.
__________________
The descent of man ~ Nixon, Friedman, Reagan, Trump.
xoxoxoBruce is offline   Reply With Quote
Old 07-21-2015, 04:04 PM   #6
Gravdigr
The Un-Tuckian
 
Join Date: Apr 2007
Location: South Central...KY that is
Posts: 39,517
Scan 'em, and tag 'em.
__________________


These statements have not been evaluated by the FDA, EPA, FBI, DEA, CDC, or FDIC. These statements are not intended to diagnose, cause, treat, cure, or prevent any disease. If you feel you have been harmed/offended by, or, disagree with any of the above statements or images, please feel free to fuck right off.
Gravdigr is offline   Reply With Quote
Reply


Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

All times are GMT -5. The time now is 06:01 PM.


Powered by: vBulletin Version 3.8.1
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.