Image to Text — Free OCR (Optical Character Recognition)

Upload a picture, screenshot, or photographed page and get its text back in seconds. ParseJet runs optical character recognition (OCR) on our own servers — no third-party AI service — and returns clean, copy-ready text in English, Chinese, Japanese, Korean, Spanish, and 20+ other languages. Use it online, or call the same OCR from the API.

Drop a file here or browse

Accepts JPG,JPEG,PNG,GIF,WEBP,TIFF,BMP files

Free — 3 requests/day, no signup. for 300 credits/month free.

How it works

Upload an image

Drop a JPG, PNG, WebP, GIF, BMP, or TIFF above — a screenshot, a phone photo of a document, or a scanned page. Up to 20 MB.

OCR runs on our servers

ParseJet detects every line of text, reads it with PP-OCR recognition models, and puts the lines back in reading order with paragraph breaks.

Copy your text

Copy the result, paste it anywhere, or automate the same conversion through the API for batches of images.

Key features

What makes this image to text stand out.

Every common image format

JPG, JPEG, PNG, GIF, BMP, WebP, and TIFF up to 20 MB — no need to convert the picture first.

20+ languages, one upload

English and every Latin-script language (Spanish, French, German, Portuguese, Polish, Turkish, Vietnamese…), Simplified and Traditional Chinese, and Japanese are read automatically. Korean, Russian and other Cyrillic languages, Arabic, Hindi, Thai, and Greek work with a one-word language hint.

Screenshots and phone photos

Tuned for screen captures and camera shots of paper — receipts, letters, whiteboards, slides. Upside-down images are corrected automatically.

Reading order kept

Lines are grouped left to right and top to bottom, with paragraph breaks where the page has them — not a bag of words.

Private by design

The OCR engine runs on ParseJet’s own servers. Your image is never sent to a third-party AI provider and is discarded as soon as the text is returned.

Same OCR via API

POST one file, get JSON back with the text and a confidence score. Python and JavaScript SDKs included.

Use cases

Common scenarios where this tool saves you time.

Screenshot to text

Pull an error message, a chat, a code snippet, or a quote out of a screenshot instead of retyping it.

Photos of paper documents

Turn a phone photo of a letter, contract, recipe, or receipt into text you can search, edit, and file.

Scanned pages and books

Digitize scans and old printouts. For multi-page scans saved as PDF, use the PDF OCR tool — it handles every page in one go.

Data entry from labels and forms

Read product labels, ID cards, and forms through the API to feed downstream systems without manual typing.

Automate with the API

Use the same tool programmatically. Works with any language — just HTTP.

cURL

# OCR a local image (JPG, PNG, WebP, GIF, BMP, TIFF)
curl -X POST https://api.parsejet.com/v1/parse/image/ocr \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "[email protected]"

# Korean, Cyrillic, Arabic, Hindi, Thai: pass a language hint
curl -X POST https://api.parsejet.com/v1/parse/image/ocr \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "[email protected]" -F "language=ko"

Python

import httpx

resp = httpx.post(
    "https://api.parsejet.com/v1/parse/image/ocr",
    headers={"Authorization": "Bearer YOUR_API_KEY"},
    files={"file": open("screenshot.png", "rb")},
)
data = resp.json()
print(data["text"])                          # the recognized text
print(data["metadata"]["ocr_confidence"])    # e.g. 0.97

JavaScript

const formData = new FormData();
formData.append("file", imageFile);      // File from an <input type="file">

const res = await fetch("https://api.parsejet.com/v1/parse/image/ocr", {
  method: "POST",
  headers: { Authorization: "Bearer YOUR_API_KEY" },
  body: formData,
});
const { text, metadata } = await res.json();
console.log(text, metadata.ocr_confidence);

Want to automate this?

ParseJet API gives you the same parsing power via a single HTTP endpoint. No ffmpeg, no poppler, no tesseract — just one API call.

curl -X POST https://api.parsejet.com/v1/parse/auto/url \
  -H "Content-Type: application/json" \
  -d '{"url":"https://example.com"}'

Read API Docs

Frequently asked questions

How do I convert a picture to text?

Drop the picture above (or click to browse), wait a second or two, and copy the text that appears. No account is needed for the first three images a day. On a phone, take the photo, open this page in the browser, and upload it from your camera roll.

What is OCR (optical character recognition)?

OCR is the technology that finds the characters in an image and turns them into real, editable text. A screenshot or a scan is just pixels — you cannot search or copy from it until OCR has read it. ParseJet uses PP-OCR deep-learning models that first locate each line of text and then recognize the characters in it.

Which languages does the OCR read?

English and all Latin-script languages (including accented letters like é, ñ, ü, ł, ş), Simplified and Traditional Chinese, and Japanese work with no settings. Add a language hint — for example ko, ru, uk, ar, fa, hi, th, or el — to read Korean, Russian and other Cyrillic languages, Arabic and Persian, Hindi, Thai, or Greek. On the localized versions of this page the hint is set for you.

Does it work with PNG and JPG screenshots?

Yes — screenshots are the best case for OCR because the text is crisp and evenly lit. PNG, JPG, and WebP screenshots from any phone or desktop are read with very high accuracy, including code and UI text.

Can it read handwriting?

Printed and typed text is where OCR shines. Clear block-letter handwriting is often readable; cursive and messy notes are hit-and-miss. Blurry, tilted, or low-light photos also reduce accuracy — retaking the picture straight-on usually fixes it.

Can I OCR a scanned PDF?

Yes. Upload the PDF to the PDF OCR tool (or the PDF to Text tool) and ParseJet detects the pages that are only images, recognizes them, and returns the whole document as text — up to 30 scanned pages per request.

Is my image uploaded to a third party?

No. OCR runs on servers ParseJet operates, using open-source recognition models. The image is processed in memory and discarded once the text is returned; it is not used to train anything.

Is it free?

Yes. You get 3 free conversions a day with no signup. A free account includes 300 credits per month (an image costs 1 credit). Paid plans start at $19/month for larger files and higher quotas.

Start extracting text for free

No signup required. Parse your first file in seconds.

View Pricing