Convert image to text
A free tool that reads the text in images with OCR (optical character recognition) and converts it into a text (TXT) file. You can open photos and screenshots, or paste an image with Ctrl+V (⌘V on a Mac). Several images can be read at once, and the text file for each image can be saved together in a ZIP. Your images are processed in your browser and never uploaded.
The first run downloads an AI model (about 17 MB) from this site; later runs use the copy stored in your browser. Your images are never uploaded.
You can also add a copied image or screenshot by pasting it with Ctrl+V (⌘V on a Mac).
Settings
How to use
- 1
Add images
Choose or drop the images to read. You can also add a copied image or screenshot by pasting it with Ctrl+V (⌘V on a Mac). You can add several images at once. Your images are never uploaded.
- 2
Choose the document language
Under "Document language", choose the language of the text in the images.
- 3
Read and save
Press "Run" to create a TXT file with the text read from the image. Save it with "Save", or copy all of it with "Copy text". For several images, use "Save all as ZIP" to save them together. You can check the recognized lines with "Check the recognized text".
About converting images to text
Text is read with PaddleOCR's text recognition model (PP-OCRv6, Apache-2.0), which runs inside your browser. The first run downloads the AI model (about 17 MB) from this site; later runs use the copy stored in your browser. On devices that cannot use the GPU (WebGPU), processing runs on the CPU and takes longer than on the GPU. Supported image formats are JPEG, PNG, WebP, AVIF, GIF (only the first frame of an animation), BMP and HEIC (iPhone photos). Long screenshots taken by scrolling can also be read.
The recognized text is arranged from the top line down, and text at the same height is joined into one line from left to right (columns and tables can come out in a different order from the reading order). Choosing Japanese under "Document language" limits recognition to the characters used in Japanese documents, and choosing Korean uses a Korean model (about 13 MB, which cannot read Chinese characters). If you only need to copy some text from a photo taken with your phone, Live Text on iPhone or Google Lens can also do it. This tool reads several images at once and saves the text of each image as a UTF-8 text file (.txt). To get the text out of a PDF, use "PDF to Text".
Specifications of this tool
- Input
- JPG, PNG, WEBP, AVIF, GIF, BMP, HEIC, HEIF
- Output
- TXT
- Limits
- No limit on the number of files (batch processing). No file size limit (processed within your device's memory). Images up to 51 megapixels and 32,767 px on the long side are read as they are (25 megapixels and 16,384 px on phones and tablets; larger images are scaled down first)
- Free, no sign-up, no app to install
- Files are never uploaded (processed in your browser)
Frequently asked questions
Are my images sent to a server?
No. Opening the images, reading the text and saving all happen in the browser on your device. Statistics that help us fix bugs (the ranges of how many images, lines and characters were read, the range of the time per image, error types and so on) are sent to this site's server. They do not include the images themselves, the recognized text or file names. You can turn the statistics off with the "Send processing statistics" switch at the bottom of the page.
Can it read text shown on my PC screen?
Yes. When you take the screenshot, select only the part you want to read (Win+Shift+S on Windows, ⌘+Control+Shift+4 on a Mac), paste it on this page with Ctrl+V (⌘V on a Mac), and press "Run". If you capture the whole screen, the text of menus at the edge of the screen can end up on the same line as the body text. You can also drop a saved image. Long screenshots taken by scrolling can also be read. Screens with light text on a dark background (dark mode) can also be read.
Can it read text in photos taken with a phone?
Yes. iPhone photos (HEIC) can be opened as they are. Images with orientation information (EXIF) are turned upright before the text is read. Photos where the text is small, blurry, shot at an angle or shaded can have more errors (this tool does not correct keystone distortion in photos). It is made for printed text, and handwritten text is not supported. It also reads vertical text, but with more errors than horizontal text, and it can miss punctuation.
Can the recognized text contain errors?
Yes. Depending on the text size, the image resolution, and blur or shadows in photos, some characters can be misread or missed. Similar-looking characters can be confused, such as the Japanese kanji numeral 「一」 and the long vowel mark 「ー」, and icons or symbols can be read as characters. When "Japanese" is chosen as the document language, characters outside the JIS level 1 and 2 kanji sets and emoji can be read as similar-looking characters. In screenshots of web pages, text inside figures or menus at the edge of the screen can end up on the same line as the body text. Press "Check the recognized text" to see the recognized lines outlined on the image, with the lines more likely to contain errors shown in a different color.
Help us improve the service
If a conversion fails, sending us the circumstances helps us track down the cause. Knowing which devices and formats fail most often, and how long processing takes, lets us revise the supported formats and fix the bugs behind them. If you don't mind, leaving this on is a real help.
What is sent
- The tool you used and the settings you chose (output format, quality, etc.)
- Your device specs (GPU, memory, core count)
- The input format with a rough size
- How many files succeeded or failed and how long it took
- The error type and details when something fails
What is never sent
- The files themselves and file names
- The recognized text and the positions of the recognized lines
- IP addresses
The data goes only to this site's own server, never to a third-party service.
For the full list, including the details each tool sends, see “Processing statistics” in the privacy policy.
Details sent by this tool (counts, times, kinds and yes/no only)
In addition to “What is sent”, this tool sends the following details.
- Image to text
- The numbers of images read, images read with no text found, images whose orientation was corrected, images read after being scaled down because they were large, images read with light and dark inverted and images whose transparent areas were filled, and of lines and characters read (all as ranges), the ranges of the share of vertical lines and of lines more likely to contain errors, the ranges of the time per image and of the AI model load time, the AI model type and version and whether it was downloaded, the range of the number of parts an image was split into to find text, and whether the GPU became unavailable.