During meetings, I often paste screenshots containing large blocks of text. It would be valuable to be able to have search include the words from those blocks of text. For instance, if I had this image in a page
It would be great if anytime that I searched “life’s work”, that image would be included.
+ 1
I do miss this funcionality too.
That is still one of the bigget advantage of OneNote agianst any other tool 
Yes. I’m taking notes from a video right now (lots of screenshots pasted right into obsidian)… I used to do this in onenote and now I can’t search the images.
If you store your vault in Dropbox, you can already search for text in images, since Dropbox has automatic OCR.
+1000 I would honestly pay monthly for OCR
There are at least two choises to get OCR working in Obsidian for now:
I haven’t tried the Tesseract installation yet, but I’ve used the Taskbone service. At least it’s easy to setup and it works well, but I understand that in some cases it’s not desirable to send private data to a third party to process.
A core plugin would be welcome, I think. I took a quick glance on the Tesseract installation instructions, and it didn’t look too difficult, but I wish a plugin could download the Tesseract binary automatically. Now you need to download it manually on Windows & Linux. On Mac you can use a brew command.
I’ve created a plugin request that we should FUND and get made: Searchable OCR
I’ll add one more. Why stop at OCR? Also use AI to determine content if these are images. see a pic of a car? Create a car tag
OCR would be a big help, since the information in my notes is often in the form of screenshots.
Just like Evernote, but less cloaked!
Just to mention… OCR for image is available in plugins. But I think that should be built in to avoid bugs and malfunction.
OCR for images and screenshots would be such a time-saver! A lot of my notes come from screenshots, and without OCR, finding text in these images is basically impossible. Having this built-in would make everything searchable and way more organized. It’d really level up Obsidian for anyone who relies on visual info! Really miss this about Apple Notes.
Reflect Notes just launched OCR for images and PDFs:
Obsidian has some catching up to do!
ChatGPT does a great job of converting pictures of hand-written notes into markdown, with the prompt “Transcribe these notes into markdown for me”. This is a powerful ability that makes it much easier to consolidate notes into obsidian markdown. More generally, access to AI image interpretation for organizing information would be a powerful tool. It would be great to have a plugin that makes use of AI image interpretation within Obsidian.
User story: I have paper notes, and I want to rapidly and neatly get that information into Obsidian. I want to take a picture of a page of my handwritten notes, and have it converted into nicely formatted markdown.
Make a plugin with camera access, that takes a picture of the notes from within Obsidian, and feeds the image into an AI model (openAI/Anthropic) with the prompt “Transcribe this text into markdown”.
Using the ChatGPT or Anthropic app to take the picture and convert to markdown, then copy and paste the markdown into Obsidian.
General ability to form Obsidian notes out of AI image interpretation without leaving the app.
My college notes have many images which are part of my notes. It would be nice if the Search function could also search inside the images, to increase keyword hits in my searches.
n/a
None. I saw some plugins which can extract text, but that is not what I am looking for. I just want the text inside images to be part of search. Preferably without having a special search command, using the normal built in search function.
Omnisearch can do that, through its own search modal.
Can make even better with a gpt/gem that will transcribe to markdown for obsidian - but also can add a yaml template everytime. what I want is some shorthand I can put in my note to add certain tags, properties without taking up too much space. like a new and improved rocketbook with more functionality.
