how do you ocr on a mac utilizing the cli or simply python at no cost

2024-01-02 12:20:56

a sort reader reached out about all of my iphone alarms and requested how particularly did I run OCR on my mac. I’m not one to gatekeep, so right here’s how one can get your self a pleasant native ocr ‘service’ that works within the terminal/CLI/python and prices 0 {dollars}:

((by the way in which sharing Shortcuts is really weird which is why I’m having you do that all manually…))
(additionally, a lot of the stuff under solely must be accomplished as soon as – after that, truly ocr’ing is simple)
open the Shortcuts app in your Functions folder
click on the “+” button to create a brand new shortcut
in the appropriate sidebar, seek for “extract textual content”

it is best to see “Extract Textual content from Picture” within the checklist of attainable actions. drag it from the appropriate sidebar into the principle space on the left
within the “Extract textual content from …” motion that’s now on the left, click on on the pale “Picture” blue-ish capsule
choose “Shortcut Enter”

you’ll see an enormous “Obtain Any enter from Nowhere” motion seem above the “Extract textual content from Shortcut Enter”. it’s all good, depart it as is
in the appropriate sidebar, seek for “copy”. it is best to see “Copy to Clipboard”. drag that motion under the “Extract textual content from …” one
- dragging the “Copy to …” motion under the “Extract from …” is absolutely annoyingly onerous. simply maintain dragging it actually under and it’ll work sooner or later
your remaining setup ought to seem like this:

within the final motion, be sure that after “Copy” it does say “Textual content from Picture”!
you’re like virtually accomplished!!
within the window title bar, sort stuff to call the shortcut. as a result of I’m from a generation, I like to recommend naming it “extract-text” or “ocr-text” or one thing easy with no areas and all lowercase. sorry. press enter after naming the shortcut in any other case it gained’t keep in mind the identify (…??)

okay chances are you’ll be truly accomplished right here I feel
check out the shortcut in your terminal:

shortcuts run ocr-text -i <A PATH TO SOME IMAGE>

then strive command-V – do you see the ensuing textual content? if sure, you’re good to go!!!
from python, you are able to do the next:

import subprocess

file_path = '... some file path ...'
ocr_out = subprocess.check_output(
    f'shortcuts run ocr-text -i "{file_path}"', shell=True
)
print(ocr_out)