Image to text
Image to text converter for photos, screenshots and scanned PDFs
Upload a photo, a screenshot or a scanned PDF and get the text on it as editable text, copied word for word. The first page is free, with no sign-up. Need the whole document? Unlock every page for $5: one payment, no subscription. A chatbot can read one clear picture, but it tends to tidy the wording and it rarely hands back a file. This tool copies exactly, page by page, and gives you TXT or Word.
- First page free, no account needed
- Whole document $5, one payment, no subscription
- Photos, screenshots, scans and multi-page scanned PDFs
- Copied word for word: nothing rephrased, corrected or translated
- Text it cannot read is marked [illegible], never made up
- Edit beside the original page, then export TXT or Word (DOCX)
What you get back
Each page comes back as plain, editable text shown next to the image of the page, so you can check one against the other. The text is a copy, not a rewrite: same words, same numbers, same language.
- The text in reading order. On a page with several columns, one column is finished before the next one starts.
- Headings on their own line and a blank line between paragraphs. Line breaks are kept where they carry meaning, such as addresses, lists and poems.
- Tables row by row, with a tab between cells, so they paste into a spreadsheet as columns.
- Numbers, dates, codes and punctuation exactly as printed. Anything unreadable appears as [illegible].
- One file for the whole document: TXT or Word (DOCX), with each page labeled.
Image to text with a chatbot, and with this tool
ChatGPT, Claude and Gemini can read text in a picture, and for one clear image that is often all you need. They are built to give a helpful answer, though, and a helpful answer is not always an exact copy. The difference shows up when the wording matters, when there are many pages, and when you need a file at the end.
| Your situation | A chatbot | This tool |
|---|---|---|
| One clear screenshot or photo | Usually reads it well | Reads it, and keeps it as a file you can export |
| The wording must be exact (a contract clause, a quote, a serial number) | Often tidies spelling, punctuation or phrasing without saying so | Instructed to copy verbatim: no corrections, no rephrasing |
| A 60-page scanned PDF | Tends to summarize, skip pages or stop partway | Reads every page separately, up to 100 pages per document |
| A page with columns or a table | May merge columns or describe the table in a sentence | Follows the reading order and keeps table cells tab-separated |
| A smudged or cut-off word | Tends to fill in something plausible | Marks it [illegible] |
| You need a document, not a chat reply | Often leaves you copying text out of the chat | Download TXT or Word (DOCX) |
How to convert an image to text
-
1
Upload the image or PDF
Drop in a photo, a screenshot, a scan or a PDF. JPG, PNG, HEIC, WebP, TIFF, GIF, BMP and PDF all work, and you can add several files to one document.
-
2
Read the first page free
The text appears next to the original page. Compare the two and fix anything in place. No account is needed for this step.
-
3
Unlock the rest and export
Happy with the first page? Unlock the whole document for $5, then download it as TXT or Word.
Why exact copying trips up general AI assistants
Reading a picture and copying a picture are two different jobs. A chat assistant looks at your image and writes a helpful reply about it. If the page has a typo, it may fix it. If a sentence is awkward, it may smooth it. If a table is long, it may give you the highlights. Each of those is reasonable behavior in a conversation and a problem when you need the page as it is.
Length makes it harder. A single screenshot is easy. A scanned PDF of 60 pages is a lot of text to reproduce in one chat reply, so assistants often shorten it, skip pages or stop partway. Then you paste pages in one at a time and stitch the answers together by hand.
This tool does one thing. Every page is read on its own with the same instruction: copy the text verbatim, in its original language, in reading order, and never invent anything. Where the page cannot be read, the text says [illegible] instead of guessing. At the end you download one file.
What people use it for
- Copying text from a screenshot, a slide or a photo of a whiteboard
- Turning a scanned PDF with no selectable text into text you can search, quote and edit
- Pulling the text out of a photographed contract, letter or form
- Getting receipts and invoices into text, with line items kept in columns
- Digitizing printed pages from books, articles and handouts
- Capturing text from signs, labels and packaging
Scanned PDFs: OCR for documents with no selectable text
A scanned PDF is a stack of pictures. You cannot select or search its text because there is none, only an image of each page. This tool turns every page of the PDF into an image and reads it, so a scan and a regular PDF are handled the same way.
What you get back is the text of each page, in page order, as TXT or Word. It does not produce a new searchable PDF and it does not rebuild the fonts, images or exact layout of the original. If you need the words, this is the right tool. If you need a pixel-perfect copy of the page, it is not.
Getting a clean result, and where the limits are
A sharp, straight image gives the best result. For a photo, lay the page flat in good light, hold the phone directly above it and fill the frame. For a screenshot, capture at full size instead of zooming out. Very small print on a very large page reads better if you crop or split it into parts.
The tool reads printed and typed text in the language it is written in and does not translate it. Photos, logos and watermarks are skipped, not described. Password-protected PDFs cannot be opened, so remove the password first. For cursive and old handwritten letters, use our handwriting to text tool, which is set up for that job. No reader is perfect, so check names, numbers and anything marked [illegible] against the page beside the text.
What it costs
First page
Free
No account. 1 document a day, or 3 a day with a free account.
Whole document
$5
One payment for one document, up to 100 pages. No subscription, never expires.
Unlimited
$29/month
Every tool on the site, up to 500 pages a month. Or $180 a year.
Questions and answers
Is this image to text converter free?
The first page of every document is free and you do not need an account. Without an account you can try 1 document a day; a free account raises that to 3 a day and lets you download the text as a TXT file.
How much does a whole document cost?
$5, paid once, for one document of up to 100 pages. There is no subscription and the unlocked document does not expire.
Can ChatGPT extract text from an image?
Yes, for a single clear image a chatbot usually reads the text well. It is less reliable when you need an exact copy, because it tends to tidy or rephrase, and on long multi-page scans, where it often summarizes or stops partway. This tool copies each page verbatim, marks unreadable text as [illegible] and gives you a TXT or Word file.
What files can I upload?
Images as JPG, PNG, HEIC (iPhone), WebP, TIFF, GIF or BMP, and PDF files with one or many pages. Scanned PDFs with no selectable text work, because every page is read as an image. Each file can be up to 25 MB and you can add up to 30 files to one document.
Does it keep the layout and tables?
It keeps the structure, not the visual design. Text follows the reading order of the page, headings and paragraphs stay separate, and tables come back row by row with a tab between cells. Fonts, colors, images and exact positions are not reproduced.
Does it read handwriting?
It is set up for printed and typed text. For cursive, old letters and handwritten notes, use the handwriting to text tool on this site, which transcribes line by line and marks every uncertain word.
Can I correct and export the text?
Yes. The text sits next to the image of the page and you can edit it in place. A free account can download TXT; an unlocked document or a Unlimited plan adds Word (DOCX).
Is my document private?
Only you can open your document. Pages are sent to an AI model provider for the sole purpose of reading them, through routes that do not use your data for training. Documents uploaded without an account are deleted after 7 days, and you can delete any document yourself at any time.
What does Unlimited include?
Unlimited costs $29 a month or $180 a year and covers every tool on the site, with up to 500 pages a month and documents of up to 300 pages. You can cancel at any time.
Updated October 20, 2026 · APPFACT STUDIO LTD