What does scanned notes to Markdown mean?
It means reading the printed words on a scanned page with OCR and writing them out as a Markdown note you can search, edit and keep.
A scanned PDF looks like text but is really a stack of pictures, so your notes app cannot search it or reflow it. OCR (optical character recognition) reads the letters in the picture. ocrtomarkdown then does the part plain OCR skips: it joins the broken lines back into paragraphs, finds the headings, rebuilds lists and tables, and drops the page numbers and running headers that repeat page after page.
What you get is the kind of note you would have typed yourself: one title, headings in order, paragraphs that flow, and text every app can search.
From scan to note in five steps
- Scan to PDFMost phones can do it. On an iPhone use Scan Documents in Notes or Files, and on Android use the Scan button in Google Drive. Keep the page flat and the light even.
- Drop the PDF aboveUp to 25 MB. Free files can run to 10 pages, so split a long chapter first, or use Pro for up to 150 pages.
- Pick your appNotion, Obsidian, Evernote, OneNote, Apple Notes or Logseq. Each gets the form it handles best.
- Check it against the scanRead the Preview tab with the original beside it. Look hardest at numbers, names and table cells, which are where OCR slips.
- Copy or downloadCopy Markdown, Copy formatted, or Download .md, then bring it into your app.
Five scan habits that change the result
The note is only as good as the picture of the page. These cost nothing.
Flat and straight
A page that curls at the spine or sits at an angle bends the lines of text. Press it flat and square it up.
Even light
Shadows across a page turn into smudges. Daylight from the side, with no hand or phone shadow, works well.
Sharp, not tiny
Fill the frame with the page. If you use a flatbed scanner, 300 dpi is a sensible setting for printed text.
One page at a time
Split a two-page book spread into two pages before you scan, so the columns do not run into each other.
Keep the edges
Do not crop away the margins. Headings and numbers often sit near the edge.
What to expect from a scan
| On the page | In your note |
|---|---|
| Printed paragraphs | Read as text and joined back into paragraphs. A clear scan gives the best result. |
| Headings | Put in order under one title, then capped to what your app handles. |
| Lists | Bullets and numbering of every kind become one proper list, nested the way the page was. |
| Tables | Simple tables come through as Markdown tables. Check the cells against the scan, especially numbers. |
| Two columns | Read in order where the converter can. Check pages that have side boxes or pull quotes. |
| Handwriting | Not promised. Handwritten notes are hit and miss, so treat any result as a rough draft. |
| Pictures, diagrams and photos | Left out. You get the text and tables around them. |
| Page numbers and running headers | Removed when the same line repeats three or more times. |
OCR is good, not perfect. Characters that look alike, such as 1 and l or 0 and O, are the usual slips, so proof the numbers.
Three ways to turn a scan into notes
| Compared on | Retyping by hand | Generic OCR | ocrtomarkdown |
|---|---|---|---|
| Effort | Retype every page | Run OCR, then clean up the output | Drop the PDF, check, paste |
| Paragraphs | As you type them | A line break per printed line | Joined back into paragraphs |
| Headings | Styled by hand | Plain text, no levels | Real headings in order |
| Tables | Rebuilt cell by cell | Rows of loose text | Markdown tables, which you check |
| Page numbers | Skipped | Scattered through the text | Removed |
| Searchable in your notes app | Yes | Yes, once cleaned up | Yes, as plain text |
Scanned notes to Markdown: your questions
Can I upload a photo instead of a PDF?
Not directly. The converter takes PDF files. Most phones turn photos into a PDF in a few taps: Scan Documents in Notes or Files on an iPhone, or the Scan button in Google Drive on Android.
Does handwriting work?
It is not promised. This site is built for printed text: typed handouts, book chapters, printed slides and recipes. Neat handwriting may come through in part, but treat any handwritten result as a rough draft.
How do I know the scan was read correctly?
You do not need to take it on trust. Keep the scan open beside the Preview tab and check numbers, names and table cells. Those are the places where OCR makes mistakes.
Why are some words wrong?
Blur, shadow, a skewed page or a very small font make letters hard to tell apart. Re-scanning flat, square and in even light usually fixes it faster than correcting by hand.
Will it keep the layout of the page?
No, and that is on purpose. It rebuilds reading order and structure (headings, lists, tables) and not the look of the page, so the note flows in any app and on any screen.
My chapter is longer than 10 pages. What do I do?
Free files run up to 10 pages. Split the PDF into parts and convert each, or use Pro for up to 150 pages per file. Each part counts as a file against your 3 a day.
Does a scanned PDF use more of my allowance?
A scanned PDF counts as one file, like any other. Free gives you 3 files a day with no account.
Can I search the result?
Yes. The result is plain text, so the search in Notion, Obsidian, Evernote, OneNote, Apple Notes or Logseq finds every word, which a scan of a page cannot offer.
What happens to my scan?
Files are processed in memory and deleted after conversion. A page that needs a second look may be sent to Google’s Gemini API. The privacy page has the details.
What does it cost?
3 files a day, up to 10 pages each, are free with no account. Pro is $6 a month for 150 files a month, files up to 150 pages.