Make your images searchable in Notion!
Notion is a great alternative to Evernote, but one feature that is still greatly missed, is that the text in your images is searchable as well. Notion promises to do this someday, but it's not there yet, so this is the alternative.
This tool allows you to add the words add_ocr
below any image in your notion account and it will try to extract the text from the image and put it below the image. Thus making the text searchable.
You'll need tesseract, as it is used for doing the actual ocr in the image.
brew install tesseract
Without brew check installing tesseract.
sudo apt install tesseract-ocr
Without apt check installing tesseract.
Make sure tesseract
works as a command in the terminal!
With Homebrew:
brew install yannick-cw/homebrew-tap/notion-ocr
With wget:
wget https://github.com/yannick-cw/notion-ocr/releases/download/0.1.5/notion-ocr_mac.zip \
&& unzip notion-ocr_mac.zip \
&& chmod +x notion-ocr \
&& ./notion-ocr --help
You should see this now
notion-ocr - enrich images with the text in the image
...
You can run it now from the current directory with
./notion-ocr-0.1_mac/bin/notion-ocr
Or just get the latest release from releases.
You can also move the executable into your PATH
to make running it more convenient.
With wget:
wget https://github.com/yannick-cw/notion-ocr/releases/download/0.1.5/notion-ocr_x86_64-linux.zip \
&& unzip notion-ocr_x86_64-linux.zip \
&& chmod +x notion-ocr \
&& ./notion-ocr --help
You should see this now
notion-ocr - enrich images with the text in the image
...
You can run it now from the current directory with
./notion-ocr
Or just get the latest release from releases.
You can also move the executable into your PATH
to make running it more convenient.
token_v2
token from https://www.notion.so/notion-ocr
with the tokennotion-ocr -t "token_here" -s 10
or if not in your path ./notion-ocr -t "token_here" -s 10
-s
option specifies to scan every x
minutes and can also be left out to run just onceadd_ocr
, the next time the tool runs, it replaces that with the text scanned from the image.tesseract
? Is it executable from the terminal?add_ocr
exactly in the line below the image?Failed loading language
engyou might need to install the tesseract eng language pack
, e.g. on Arch I needed to run pacman -S tesseract-data-eng